The Evolution of Inbreeding in Western Redcedar (Thuja Plicata: Cupressaceae)
THE EVOLUTION OF INBREEDING IN WESTERN REDCEDAR (THUJA PLICATA: CUPRESSACEAE)
by
LISA MARIE O'CONNELL B.A. University of Ottawa, 1993 B.Sc. Dalhousie University, 1995 M.Sc. Queen's University, 1997
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES
(Department of Forest Sciences)
We accept this thesis as conforming to the required standard
THE UNIVERSITY OF BRITISH COLUMBIA 2003 © Lisa Marie O'Connell, 2003 In presenting this thesis in partial fulfilment of the requirements for an advanced
degree at the University of British Columbia, I agree that the Library shall make it
freely available for reference and study. I further agree that permission for extensive
copying of this thesis for scholarly purposes may be granted by the head of my
department or by his or her representatives. It is understood that copying or
publication of this thesis for financial gain shall not be allowed without my written
permission.
Department of forfs't Sci e rt c*5
The University of British Columbia Vancouver, Canada
Date April H , 2^003
DE-6 (2/88) Abstract
Long-lived woody plants usually show high levels of outcrossing, inbreeding depression and genetic diversity compared to other plants. A review of the literature showed a mean oucrossing rate of 83.5 in conifers, and a positive, but weak, correlation between outcrossing and genetic diversity. Among conifers, western redcedar (Thuja plicata, Cupressaceae) has one of the highest rates of self-fertilization and lowest amount of genetic diversity, and thus offers the opportunity to study the evolution of inbreeding in a predominantly outcrossing group of plants.
This thesis links the evolution of inbreeding in redcedar with a loss in inbreeding depression and genetic diversity. Using one polymorphic isozyme marker, I obtained an average population outcrossing estimate of 71% over six natural populations of redcedar. I developed 13 highly polymorphic microsatellite markers to conduct a finer-scale study of the mating system and genetic structure of redcedar. A new method of bulking seedlings to estimate outcrossing rates was used to identify ecological correlates of outcrossing. Selfing rates increased significantly with tree height in four different populations. Pollen from larger trees probably made up a larger proportion of the surrounding pollen cloud, increasing self-pollination. There was no variation, however, in the amount of inbreeding among crown positions within trees. In a seed orchard, a combination of controlled crosses and isozyme markers showed evidence that post-pollination competition between embryos within an ovule decreased selfing. I used eight microsatellite loci to study patterns of range-wide genetic structure in redcedar. A phylogeographic analysis suggests that redcedar probably survived in three separate refugia during the last glaciation.
These results also suggest that if a species-wide bottleneck is at the root of reduced genetic diversity in redcedar, it probably predates the last glaciation. The combination of an inbreeding mode of reproduction and a bottleneck probably contributed to the decrease in genetic diversity presently observed in redcedar. Finally, after screening 80 trees at eight microsatellite loci, a single stepwise mutation was observed, yielding a somatic mutation rate of 6.3 x 10"4 (95% CI:
3.0 x 10"5 - 4.0 x 10"3) mutations per locus per generation in western redcedar. iv
Table of Contents
Abstract ii Table of Contents iv List of Tables viii
List of Figures x
List of Appendices • X1i
Acknowledgments XU1
Published papers X1V
Chapter 1 General introduction and overview 1 The evolution of plant mating systems 1 Conifers 3 Patterns of genetic diversity in conifers 3 Review of outcrossing rates in conifers 6 Outcrossing rate and genetic diversity 8 Natural populations vs seed orchards 10 Inbreeding depression in conifers 11 Mating system of conifers 11 The genus Thuja (Cupressaceae) 12 Western Red Cedar (Thuja plicata) 12 Ecology of Thuja plicata - 12 Genetic diversity in Thuja plicata 13 Inbreeding depression in Thuja plicata 13 Species-wide bottleneck 14 Thesis overview 15
Chapter 2 The mating system in natural populations of western redcedar 16 Introduction 16 Materials and methods IV Sample collections IV Isozyme analyses 18 Data analysis 18 Results 19 Discussion 20 Outcrossing rates 20 Factors affecting mating systems 20 Correlation of paternity 22 V
Chapter 3 Characterization of microsatellite loci in western redcedar 24 Introduction 24 Materials and methods 24 Clone development 24 Screening for polymorphisms 25 Results and Discussion 26
Chapter 4 Fine-scale estimation of outcrossing in western redcedar with microsatellite assay of bulked DNA 29 Introduction 29 Materials and methods 31 Bulking tests 31 Sample collections 32 DNA assay of bulks 33 Estimation of outcrossing from bulk samples 34 Results 36 Allele detection 36 Bulking tests 37 Genetic diversity 39 Outcrossing rates 39 Discussion 43 Variation in outcrossing rates 43 Population outcrossing rates 44 Bulking samples 45
Chapter 5 Polyembryony and early inbreeding depression in a self-fertile conifer, Thuja plicata (Cupressaceae) 46 Introduction 46 Materials and Methods 48 Pollinations 48 Seed viability 50 Embryo competition 51 Expected seed set and selfing with polyembryony 52 Fitness of self-pollen 54 Results 56 Seed set 56 Realized selfing rates 60 Success of self pollen 62 vi
Discussion 63 Polyembryony as a rescue mechanism 63 Embryo competition 64 Early inbreeding depression self-incompatibility 65 Purging of inbreeding depression 65 The importance of pre-pollination mechanisms 66
Chapter 6 Range-wide genetic structure and diversity in western redcedar 68 Introduction 68 Matherials and methods 70 Sample collection 70 Microsatellite screening 74 Diversity analyses 74 Phylogeography 75 Clines in allele frequencies 76 Linkage disequilibrium 76 Bottleneck test 76 Results 78 Genetic Diversity 78 Genetic Structure 80 Isolation by distance 82 Northern vs southern populations 84 Mating system 85 Allele size distribution 85 Clines in allele frequencies 89 Bottleneck test 92 Discussion 95 Phylogeographic structure 95 Population differentiation 99 Reduction in genetic diversity 100 Timing a species-wide bottleneck 101 Inbreeding and genetic diversity 102
Chapter 7 Somatic mutations at microsatellite loci in western redcedar 104 Introduction 104 Materials and methods 105 Estimating mutation rate 105 Sample collections 107 vii
Microsatellites 108 Results 109 Microsatellite mutations 109 Type of mutation 110 Somatic mutation rate estimate 110 Discussion Ul Somatic mutation rate Ul Mutation model 112 The consequences of somatic mutations in redcedar 112 Genetic mosaicism 113
Chapter 8 General discussion and conclusions 114 Main findings 114 Mating system 114 Inbreeding depression 115 Genetic structure and diversity 117 Tying it all together 118 Further research 119 Review of mating systems in trees 119 Glacial refugia 119 Allele distribution 120 Mating system at the edge of the distribution 121 Related species 121
References 123 Vlll
List of Tables
1.1 Mean levels of within population isozyme variation in three species of Thuja compared to mean levels in gymnosperms and other plants 5
1.2 Mean outcrossing rate (t ± SD) by genus in 52 species of conifers 6
1.3 Outcrossing rates in 13 species of conifers with estimates from both natural and seed orchard populations 10
2.1 Locations, sample sizes (AO, gene frequency of the most common allele at locus G6pd,
population outcrossing rates (0 and correlation of paternity (rp) of Thuja plicata populations in southwestern British Columbia 23
3.1 Characterization of Thuja plicata microsatellites in a coastal and two interior populations ...
27
4.1 Probabilities of band patterns observed for a single progeny, and for bulked progenies of sizes 2 and 3, conditioned upon maternal genotype (homozygous A,A, or heterozygous
A,A,) 35
4.2 The total number of alleles detected in samples bulked before DNA extraction in two trees of Thuja plicata. Samples were scored at four microsatellite loci (TP1, TP3, TP9 and TP11) 38
4.3 Genetic diversity measures and total number of alleles detected at four microsatellite loci (TP1, TP3, TP9 and TP11) in four natural populations of Thuja plicata 40
4.4 Outcrossing rates (SE) at different positions within the crown of trees in four natural populations of Thuja plicata 41
4.5 Mean tree heights and individual tree outcrossing rates (0 in four populations of Thuja plicata 43
5.1 Experimental design of a pollination experiment in four trees in a Thuja plicata seed orchard. One hundred percent self-pollen (selfed) and 0% self-pollen (crossed) treatments as well as pollen mixtures with three different ratios of self/cross pollen (25%/75%; 50%/50%; 75%/25%) were applied to each tree 49
5.2 Probabilities of setting a full seed (/•) and setting a selfed seed (s,) with one, two or three embryos per ovule (n) 53
5.3 The proportion of full seeds (SE) and inbreeding depression at the seed stage in four western redcedar trees (181, 295, 431 and 432) 58 ix
5.4 The proportion of full seeds expected (/•) based on the number of embryos per ovule (n)
and the proportion of self pollen applied (pk) in four Thuja plicata trees with varying levels of embryo viability 59
5.5 The proportion full seeds expected (j)) without polyembryony (when n = 1) and observed full seeds for three different proportions of self-pollen in four Thuja plicata trees 60
5.6 The proportion of selfed seeds expected (when n = 1) and observed for three different proportions of self-pollen in three Thuja plicata trees 62
5.7 Fitness of self-pollen relative to outcross-pollen (ws) when applied at different proportions
in three Thuja plicata trees 63
6.1 Location of 23 sampled populations of Thuja plicata '. 73
6.2 Mean diversity at eight microsatellite loci in 23 populations of Thuja plicata 75
6.3 Genetic diversity measures in 23 populations of Thuja plicata at seven microsatellite loci...
.' 79
6.4 F-statistics at eight microsatellite loci in Thuja plicata 80
6.5 Analysis of molecular variance (AMOVA) of the effects of populations and groups (North vs South) on the distribution of genetic diversity in Thuja plicata based on seven microsatellite loci 82
6.6 Analysis of covariance (ANCOVA) of the effects of geographical distance between populations and groups (N vs N, S vs S, and N vs S) on pairwise genetic distance (6).... 84
6.7 Alleles with frequencies significantly correlated with latitude at the 0.05 level at seven microsatellite loci in Thuja plicata ....90
6.8 Results of the Bottleneck test for two mutation models in Thuja plicata. Twenty-three populations were divided into three groups based on the phylogeographic analysis: southern populations (excluding California), northern populations and California only. Isozyme data for eight populations were obtained from Yeh (1988) 94
7.1 Description of eight microsatellite loci used to genotype 80 Thuja plicata trees from four natural populations 108 X
List of Figures
1.1 Traits that are positively correlated in plants in meta-analyses of the literature 2
1.2 Distribution of outcrossing rates in 52 species of conifers 7
1.3 Correlation between outcrossing rate (?) and mean expected heterozygosity (Hep) in 40 species of conifers. Species of Thuja are indicated by crosses 9
4.1 Diagram of six cone collection positions in a Thuja plicata tree: (1) top, vigorous branches (2) top, inner branches (3) mid, outer branches (4) mid, inner branches (5) lower, outer branches and (6) lower, inner branches 33
4.2 Band intensity profile of three bulked individuals at locus TP9 from lane 6 on the microsatellite gel. The four detected alleles are indicated by black arrows on the band intensity profile 37
4.3 Individual tree outcrossing rate estimates regressed on tree height in four populations of Thuja plicata. N = 73 42
5.1 The proportion of selfed seeds expected with different proportion of self-pollen and number of embryos (n) within an ovule. Two different outcomes are shown: the outcrossed embryo always outcompetes the selfed embryo (outcross wins) or self and outcross embryos are equally competitive (chance). All embryos are viable and there is no inbreeding depression 55
5.2 The proportion of full seeds obtained in four Thuja plicata trees with different proportions of self-pollen applied. ./V = 495 57
5.3 The proportion of selfed seeds expected and observed (± SE) with different proportion of self-pollen. N = 2028. n, number of embryos per ovule. Embryo viabilities for the expected selfed seeds is based on the mean of three trees 61
6.1 Range map and location of 23 sampled populations of Thuja plicata. The shaded areas indicate the range of western redcedar 72
6.2 Neighbor-Joining tree of 23 populations of Thuja plicata based on Nei's standard genetic distance 81
6.3 Pairwise genetic distance (0) as a function of geographic distance between populations of Thuja plicata 83
6.4 Average number of alleles per locus (A/L), mean expected heterozygosity (Hep) and mean inbreeding coefficients (F) in 23 populations of Thuja plicata as a function of latitude.. 86 xi
6.5 a-d Allele size distribution over the range of western redcedar at four microsatellite loci: TPl,TP3,TP4andTP6 : 87
6.5 e-h Allele size distribution over the range of western redcedar at four microsatellite loci: TP7, TP8, TP9 and TP11 88
6.6 Relative frequency of (a) 12 northern and (b) 23 southern alleles as a function of latitude in 23 populations of Thuja plicata 91
6.7 Proportion of alleles at eight loci in different frequency classes for southern populations and northern populations of Thuja plicata. (n = 189 alleles) 93
6.8 Hypothesized post-glacial colonization routes for Thuja plicata from three glacial refugia discussed in the text: (1) California, (2) Queen Charlotte Islands, and either (3) western Oregon or (4) northern Idaho 97
7.1 Image of a microsatellite gel showing the genotype at locus TP9 for two different heights within the same tree. Two collections of ten bulked megagametophytes were made from three heights in each tree 109
7.2 Allele distribution at locus TP9 over four populations of Thuja plicata (N = 80 trees). The new allele, which increased from 34 to 35 dinucleotide repeats, is indicated by the white box, with the arrow showing the original allele 110 Xll
List of Appendices
I Description of genetic diversity parameters 143
II Literature review of genetic diversity in 50 species of conifers including 35 species in the Pinaceae, 13 in the Cupressaceae and one in the Taxodiaceae 144
III Literature review of outcrossing rates in 52 species of conifers and 32 species of
angiosperm trees 149
IV Western redcedar foliage DNA CTAB extraction protocol 156
V Twelve western redcedar DNA sequences from which primer pairs where designed and amplified scorable and variable microsatellite loci 158
VI Pairwise genetic distances between 23 populations of Thuja plicata based on eight
microsatellite loci. Nei's (1972) genetic distance is above the diagonal.and Fsl (0) below the diagonal 162 Xlll
Acknowledgements
First of all, I thank my supervisor Kermit Ritland whose original way of thinking and his receptiveness to new ideas helped approach science in a new an interesting way. I thank the members of my supervisory committee John Russell, Sally Otto and Yousry El-Kassaby, who have not only kept me on the right track but also helped me tremendously along the way with their words of encouragement. And of course the lab work would never have gone so smoothly if Carol Ritland hadn't been there to keep everything running.
I thank Frederique Viard who worked out the isozymes protocols for redcedar and began the work, Jeff Glaubitz and Gwenael Vourc'h who provided samples of redcedar DNA. I thank
John Russell for showing me the ropes in Cupressaceae research. He provided thousands of samples, set up the facilities and provided the information needed to conduct my studies. I thank
Heidi Collinson and Tim Crowder for their help at the Mt. Newton Seed Orchard.
I thank my parents, Rod and Rolande, who always supported and encouraged me through my many, many, many years of university and made sure I would get through it. I also thank my lab "family" for their friendship and support: Bryan Ie, Dawn Marshall, Carol Goodwillie, Dilara
Ally, Charles "Chin-Lin" Chen, Marissa LeBlanc, Yanik Berube, Hugh Wellman, Washington
Gapare, Jodie Krakowski, Allyson MisCampbell, Dawn Cooper, Mark Van Kleunen, Jaclyn
Beland and Jennifer Wilkin. We created a home far away from home and were always there for each other. Thank you to Mark, Carol, Allyson, Dilara, Marissa, Jodie for taking the time to read this thesis and suggesting many improvements. I also thank Jeannette Whitton, Sally Aitken and
Dan Shoen for their careful reading and helpful comments on this thesis.
Funding was provided by a Natural Sciences and Engineering Research Council of
Canada post-graduate scholarship (PGSB) and an Isaak Walton Killam pre-doctoral Fellowship and a research assistanceship from K. Ritland.
"After great pain, a formal feeling comes." Emily Dickinson xiv
Published Papers
Chapter 2 is a revised version of the following paper: O'Connell, L. M., F. Viard, J. Russell, and
K. Ritland. 2001. The mating system in natural populations of western redcedar (Thuja plicata).
Canadian Journal of Botany 79: 753-756.
For this study Frederique Viard collected the seeds from two populations, optimized the
isozyme protocol and genotyped the seedlings for the first year of the study. John Russell collected the seeds from five populations during the second year. I genotyped all the seedlings
from the second year, conducted the analyses and wrote the paper. Kermit Ritland supervised
the study and the analyses, and edited the manuscript.
Kermit Ritland ...
Chapter 3 is a revised version of the following paper: O'Connell, L. M., and C. E. Ritland. 2000.
Characterization of microsatellite loci in western redcedar (Thuja plicata). Molecular Ecology 9:
1920-1922.
The clones containing microsatellites were obtained from Craig Newton (BC Research).
Carol Ritland optimized the protocol for sequencing the clones and supervised all the steps from
primer design, DNA isolation and microsatellite screening, and edited the paper. I conducted the
majority of the lab work and wrote the paper.
Carol Ritland: .... 1
Chapter 1
General introduction and overview
The evolution of plant mating systems
A major trend in the evolution of plant mating systems is a transition from cross- fertilization to self-fertilization. Identifying the selective factors involved in the evolution of plant mating systems has been the subject of a large amount of both theoretical (e.g.: Lande and
Schemske, 1985; Jarne and Charlesworth, 1993; Uyenoyama et ah, 1993) and empirical work resulting in estimates of outcrossing rates for over 200 species of plants (reviewed in Barrett et al., 1996). Meta-analyses of the literature have shown that a species' mating system, genetic diversity, inbreeding depression and life history are interdependent (Fig. 1.1). Species with high self-fertilization rates tend to show lower genetic diversity and less inbreeding depression than predominantly outcrossing species (Charlesworth and Charlesworth, 1995; Hamrick and Godt,
1996; Husband and Schemske, 1996). Annual plants are often selfers, and long-lived woody plants are predominantly outcrossers (Barrett and Eckert, 1990; Barrett et al, 1996). Long-lived woody plants also tend to have higher genetic diversity than other plants (Hamrick et ah, 1992;
Hamrick and Godt, 1996).
Inbreeding depression is seen as a major driving force in the evolution of mating systems, favoring many traits that prevent plants from self-fertilizing (Charlesworth and Charlesworth,
1987). Longer-lived plants, including woody plants, are expected to accumulate a higher genetic load through somatic mutations, consequently maintaining outcrossing (Morgan, 2001). Highly deleterious recessive mutations affecting early life stages such as seed production, germination and early survival are expected to be readily purged in selfing plants while later acting inbreeding depression affecting growth and fertility should be more difficult to purge (Husband and Schemske, 1996). In a review comparing the timing of inbreeding depression and the mating system of plants, Husband and Schemske (1996) found that selfing plants had less inbreeding depression during the early stages of their life cycle than did outcrossers. However, 2
at later stages there was no difference in inbreeding depression between self-fertilizing and outcrossing species. Short-lived plants may also be more successful at purging inbreeding depression than longer-lived species but overall the results are inconclusive (Byers and Waller,
1999). Nevertheless, most tropical angiosperm trees and temperate conifers show highly reduced seed set following self-pollination (Bawa, 1974; Kormut'ak and Lindgren, 1996; Husband and
Schemske, 1996).
Genetic Outcrossing diversity 4
Inbreeding Lifespan depression f
Fig. 1.1 Traits that are positively correlated in plants in meta-analyses of the literature. References (a) Hamrick and Godt 1992 (b) Husband and Schemske, 1996 (c) Barrett and Eckert, 1990 (d) Barrett et al, 1996 (e) Hamrick and Godt, 1996 (f) Byers and Waller, 1999.
Although inbreeding depression is important in shaping selection on mating systems, other theoretical models have also incorporated pollen ecology and life-history (Holsinger, 1991;
Morgan et al, 1997; Johnston, 1998). Different costs and benefits are associated with both selfing and outcrossing. For example, the advantage of reproductive assurance through selfing in 3
annual and colonizing plants in the absence of outcross pollen has long been recognized
(Stebbins, 1950). In perennial plants, reduced survivorship and fecundity in later years may increase the cost of selfing by producing less fit inbred seeds, rather than maximizing fitness by delaying reproduction until non-inbred pollen is available (Morgan et al, 1997). The relative amounts of self and outcross pollen available to a plant will also affect selection on its mating system (Holsinger, 1991). For example, the large size of trees can increase self-pollination through geitonogamy (self-pollination between different flowers on the same plant) and consequently lead to the evolution of mechanisms to prevent selfing (Barrett et al, 1996).
Variation in space and time in the availability of unrelated pollen can lead to a stable mixed- mating system (Holsinger, 1991). Although traits correlated with mating systems can be identified at the species level, a large amount of the variation in outcrossing rates occurs within species. To better identify traits associated with plant mating systems, differences among closely related species, or among populations and individuals within a species, need to be examined
(Barrett and Eckert, 1990; Barrett et al, 1996).
In this thesis I will obtain estimates of self-fertilization, early inbreeding depression and genetic diversity in a conifer, western redcedar (Thuja plicata Donn ex D. Don, Cupressaceae), using two classes of genetic markers: isozymes and microsatellites. I will show how the history of western redcedar has shaped the evolution of these three linked quantities. To set the stage for my study I will first review patterns of genetic diversity and mating systems in conifers and other trees. I will then present data from previous studies on the genetic diversity, mating system, ecology and evolutionary history of Thuja plicata and closely related species.
Conifers
Patterns of genetic diversity in conifers - In general, conifers have higher genetic diversity than other plants at the population level, as measured by the proportion of polymorphic 4
loci (PPL), alleles per locus (A/L) and average expected heterozygosity (Hep \ Table 1.1; genetic
diversity parameters are described in Appendix I). In conifers, genetic variation resides mostly
within, rather than between, populations. At the species level, Hardy-Weinberg expected
heterozygosity (Hes, Weir, 1990) is not significantly higher for gymnosperms (mostly
conifers)(//„ = 0.169) than for either monocots (Hes = 0.159) or dicots (Hes = 0.184; Hamrick and
Godt, 1996). The total genetic diversity residing among populations is significantly lower in
gymnosperms (G,, = 0.073) than in other plants (monocots: Gsl = 0.157; dicots: Gs, = 0.184;
Hamrick and Godt, 1996). I gathered data on mean levels of population genetic diversity in
conifers from 68 isozyme studies (Appendix II). Because most studies sampled only part of a
species' range, I decided to only report mean population measures of genetic diversity (Hep,
PPL, A/L) rather than diversity at the species level (Hes). Measures of within population genetic
diversity were lower for 13 species of Cupressaceae, than for 35 species of Pinaceae (Hep: t - test
= 2.08, n = 49, P = 0.043; PPL: t - test = 2.18, n = 40, P = 0.036; and A/L: t - test = 2.15, n = 40,
P = 0.038; Table 1.1). Table 1.1 Mean levels of within population isozyme variation in three species of Thuja compared to mean levels in gymnosperms and other plants.
Species POP L PPL A/L Hep Reference
T. plicata 49 trees 9 0 1 0 Copes, 1981
T. plicata 8 15 15.8 1.17 0.039 Yeh, 1988
T. plicata 1 9 12 1.22 0.04 El-Kassaby et al, 1994
T. occidentalis 6 18 37 1.5 0.094 Perry et al, 1990
T. occidentalis 6 " 11 13.9 1.17 0.034 Matthes-Sears et al, 1991
T. occidentalis 6 20 54.2 1.6 0.129 Lamy et al, 1999
T. orientalis 14 26 57 1.89 0.144 Xie et al, 1992
Conifers /V = 50 10.0 19.8 50.4 1.72 0.154 Appendix II
Pinaceae N -36 10.7 20.8 54.0 1.78 0.163 Appendix II
Cupressaceae N =13 7.4 17.1 42.3 1.59 0.124 Appendix II
Gymnosperms N = 102 8.9 17.3 53.4 1.83 0.151 Hamrick et al, 1992
All plants N = 669 12.3 17.3 34.6 1.52 0.113 Hamrick et al, 1992
N, number of species sampled; POP, Number of populations sampled; L, Number of loci
sampled; PPL, percent polymorphic loci; A/L, mean number of alleles per locus; Hep, mean expected heterozygosity within populations. See Appendix I for more details. 6
Review of outcrossing rates in conifers - Conifers are predominantly outcrossing, however several species show moderate levels of selfing (the proportion of self-fertilized offspring). I gathered data from more than 100 studies estimating the outcrossing rate in both natural and seed orchard populations of conifers and angiosperm trees (Appendix III). Table 1.2 summarizes the outcrossing rates for a total of 52 species of conifers from eight genera and three families. The majority of species for which data were available were in the Pinaceae, specifically in the genus Pinus. The average outcrossing rate for all conifer species was 0.835 ±
0.171 SD. While most species of conifers have outcrossing rates of over 80%, a quarter of the species are below this value and show significant amounts of inbreeding (Fig. 1.2). One species,
Chihuahua spruce (Picea chihuahuana) is almost entirely inbred, but this species is limited to a few small isolated populations (Ledig et al., 1997). The mean outcrossing rate for 32 angiosperm tree species is not significantly different than for conifers (mean t = 0.896 ± 0.164
SD; t - test = 1.619, df = 82, 2-tailed P = 0.109; data in Appendix III.)
Table 1.2 Mean population outcrossing rate (t ± SD) by genus in 52 species of conifers (see Appendix III).
Genus Number of species Mean t ± SD Family
Abies 5 0.889 ± 0.066 Pinaceae
Larix 4 0.821 ± 0.062 Pinaceae
Picea 9 0.732 ± 0.285 Pinaceae
Pinus 28 0.878 ±0.124 Pinaceae
Pseudotsuga 1 0.880 Pinaceae
Tsuga 1 0.975 Pinaceae
Thuja 3 0.597 ±0.135 Cupressaceae
Cunninghamia 1 0.902 Taxodiaceae 7
14 • Thuja spp. • Other conifers 12 0 Picea chihuahuana
10
Number 8 of species 6
2
0 0 0.2 0.4 0.6 0.8 Outcrossing rate
Fig. 1.2 Distribution of outcrossing rates in 52 species of conifers. Data are in Appendix III. 8
Outcrossing rate and genetic diversity: Estimates of both genetic diversity and outcrossing rates were available for 40 species of conifers (Appendices II and III). A species'
outcrossing rate and genetic diversity (Hep) were positively, but weakly, correlated (r = 0.406, N
= 40, P = 0.009; Fig. 1.3). Species with a high selfing rate showed reduced genetic diversity; however, several species with low genetic diversity still maintained high outcrossing rates. The correlation remained significant when the outlier, Picea chihuahuana, was excluded from the analysis (r = 0.387, N = 38, P = 0.015). These results suggest that high levels of inbreeding can reduce levels of genetic diversity at the population level in conifers. In species with higher outcrossing rates, however, other factors may be contributing to a reduction of genetic diversity.
Other species of conifers with low levels of genetic diversity may also show a link between inbreeding and a reduction in genetic diversity. Two species of conifers that show almost no isozyme variation, red pine (Pinus resinosa) and Torrey pine (Pinus torreyana), were not included in this analysis because no estimates of outcrossing were available. In red pine, only
four polymorphic loci have been observed out of 64 sampled (Hep - 0.002, based on 27 enzyme systems; Fowler and Morris, 1977; Allendorf et al, 1982; Simon et al, 1986; Mosseler et al,
1991). The lack of early inbreeding depression in 46 red pine trees following controlled selfed
(71% seed set) vs outcrossed (72% seed set) pollinations, suggests that selfing is potentially high in natural populations (Fowler, 1965). Mating system information lacking for Torrey pine as
well, which exists in only two populations showing no isozyme variation (Hep - 0 based on 59 loci; Ledig and Conkle, 1983; although the two populations are fixed for different alleles at one locus). 0.30
0.25 I
0.20
T^vrientaUs 0.15 ^ • x
0 10 - Picea chihuahuana • :?xgr;occidentals
0.05 - x T. plicata
0.00 0,00 0.20 0.40 0.60 0.80 1.00
©ukrassirigfrate
Fig. 1.3 Correlation between outcrossing rate (0 and mean expected heterozygosity (Hep) in species of conifers. Species of Thuja axe indicated by crosses. 10
Natural populations vs seed orchards: Data on outcrossing rates from both natural and seed orchard populations were available for 13 species of conifers (Table 1.3). Because seed orchards are designed to minimize self-pollination, outcrossing rates are expected to be higher in orchards than in natural populations (Adams and Birkes, 1991). Indeed, ten of the 13 species of conifers showed a higher outcrossing rate in seed orchards compared to natural populations. But overall there was no significant difference in outcrossing rates between the two population types
(Wilcoxon sign rank test: T = 19.5, N= 13, l-tailed P = 0.092).
Table 1.3 Outcrossing rates in 13 species of conifers with estimates from both natural and seed orchard populations. References are in Appendix III.
Population type: Natural Orchard
Abies procera 0.94 1
Larix decidua 0.809 0.852
Larix occidentalis 0.894 0.803
Picea abies 0.895 0.937
Picea glauca 0.855 0.931
Picea mariana 0.827 0.837
Picea omorika 0.84 1
Pinus caribaea 0.921 1
Pinus leucodermis 0.802 0.86
Pinus sylvestris 0.965 0.975
Pinus tabulaeformis 0.864 0.957
Pseudotsuga menziesii 0.886 0.874
Thuja plicata 0.715 0.32
Average = 0.867 ± 0.016 SE 0.873 ± 0.050 SE 11
Inbreeding depression in conifers - Inbreeding depression (8) in conifers can be
expressed through reduced seed viability, reduced germination, slower growth rate and higher
mortality following self-fertilization. Conifers usually show strong inbreeding depression
expressed mostly at early life stages (Charlesworth and Charlesworth, 1987; Sorensen, 1999).
The proportion of full seeds after self-pollination (ws) relative to full seeds cross-pollination (wx)
in 17 species of conifers (all in the Pinaceae) averaged 39% (5 = 1 - ws I wx= 0.61, reviewed in
Kormut'ak and Lindgren, 1996). In 10 species of outcrossing conifers, inbreeding depression at
the seed stage (5 = 0.58) was much higher than during germination (8 = 0.09) or during growth
and reproduction (8 = 0.18) (reviewed in Husband and Schemske, 1996).
Mating system of conifers - In conifers, pre-pollination mechanisms such as monoecy
(separate male and female cones) and dichogamy (separation in time between pollen shedding
and female receptivity) can reduce self-pollination (Richards, 1986). Unlike angiosperms,
gymnosperms seem to lack early self-incompatibility mechanisms occurring between pollination
and fertilization. However, polyembryony, the presence of several embryos within an ovule, can potentially decrease the number of inbred progeny in conifers. Female gametophytes can possess several archegonia with the same haploid maternal genotype, but fertilized by different pollen parents. Several embryos begin to develop within an ovule but only one embryo eventually survives. If both outcrossed and selfed embryos are viable, outcrossed embryos could
outcompete less fit inbred embryos within the same ovule, decreasing the proportion of self- fertilized progeny (Sorensen, 1982; Savolainen et al, 1992). Polyembryony can also potentially decrease the number of empty seeds in species with high inbreeding depression. If both selfed
and outcrossed embryos are found within the same ovule, the death of an inviable selfed embryo
will not necessarily cause the abortion of a seed if a viable embryo is also present (Sorensen,
1982). 12
The genus Thuja (Cupressaceae)
There are six species in the genus Thuja worldwide, two in North America (T. plicata and
T. occidentalis) and four in eastern Asia (T. koriaensis, T. orientalis, T. standishii and T.
sutchuensis; Vidakovic, 1991). The population genetic structure of three of these species (T. plicata, T. occidentalis and T. orientalis) has been studied using isozymes. Mean levels of isozyme diversity in T. plicata and T. occidentalis are lower than in other conifers, while genetic
diversity in T. orientalis is similar to other species of conifers/gymnosperms (Table 1.1; Fig.
1.3). All three species of Thuja have a mixed mating system with outcrossing rates among the
lowest in conifers (Fig. 1.2). Thuja orientalis showed an average of 75% outcrossing in natural populations (Xie et al, 1991), while Thuja occidentalis showed 64% outcrossing in Ontario populations (Perry and Knowles, 1990) and 29% outcrossing in Quebec populations (Lamy et al, 1999). The outcrossing rate in a seed orchard population of Thuja plicata was 32% (El-
Kassaby et al, 1994). Thus, the genus Thuja offers the opportunity to study the evolution of
selfing in a predominantly outcrossing group of plants, the conifers.
Western Red Cedar {Thuja plicata)
Ecology of Thuja plicata - The range of Thuja plicata Donn ex D. Don (Cupressaceae) extends along the Pacific coast of North America from southeastern Alaska to northern
California, and in the interior from east-central British Columbia into the panhandle of Idaho and western Montana. The coastal and interior parts of the range are essentially isolated from each other, but a few stands occur between the Coast Ranges and the Selkirk Mountains near the southern border of British Columbia (Minore, 1990). Reproduction in redcedar usually starts at
20-30 years but can start as early as 10 years in trees exposed to sunlight (Minore, 1983). Trees can often live over 1000 years (Minore, 1983). Vegetative reproduction by layering or rooting of fallen branches is common in mature stands in Idaho (Parker and Johnson, 1988) and may lead to 13
clonal clusters. Pure stands of T. plicata are rare, and it usually grows in mixed-species, uneven- aged stands and occurs at all stages of forest succession (Minore, 1983).
Genetic diversity in Thuja plicata - Measures of genetic diversity show low variation within and low differentiation among populations of western redcedar. First, there is low variation in relative amounts of leaf oil terpenes over the entire range of western redcedar, but some minor differences have been detected between coastal and interior populations (von
Rudloff and Lapp, 1979; von Rudloff et al., 1988). Second, Copes (1981) found no isozyme variation at nine loci in trees from Washington and Oregon, and Yeh (1988) and El-Kassaby et al. (1994) found very low variation in British Columbia populations (Table 1.1). Overall, of 21 isozyme loci studied, five showed some variation in at least one population, and only one,
Glucose-6-phosphate dehydrogenase (G6pd), was variable in all studied populations (Yeh, 1988;
El-Kassaby et al, 1994). Third, Glaubitz et al. (2000) screened individuals from throughout the range of T. plicata with RFLP (restriction fragment length polymorphism) probes and found little differentiation among regions. Finally, studies that have examined variation in phenotypic traits have found either no difference (Bower and Dunsworth, 1988) or low differentiation among populations (Rehfeldt, 1994). However, significant differences in resistance to Keithia blight, caused by the fungus Didymascella thujina, have been found among populations of redcedar (J.
Russell, pers. comm.)
Inbreeding depression in Thuja plicata - Unlike most other conifers, Thuja plicata has shown little inbreeding depression during early life stages. In one study, two self-pollinated clones set more seeds per cone than four cross-pollinated clones (Owens et al, 1990). In a review of inbreeding depression in plants, Husband and Schemske (1996) found that predominantly outcrossing conifers showed a cumulative inbreeding depression over their lifespan of 5 = 0.67 while in western redcedar, life-time inbreeding depression was only 8 = 0.30 14
(J. Russell cited in Husband and Schemske, 1996). Most of the inbreeding depression in western
redcedar occurred at the seed stage (8 = 0.28) but was much lower than in other conifers at the
same stage (8 = 0.58). More recently, inbreeding depression of 10% in growth rate after 11 years
has also been measured in western redcedar (J. Russell, pers. comm.)
Species-wide bottleneck - A reduction in a species' genetic diversity can occur through a
reduction in the effective population size, either through a large historical bottleneck or small
recurring bottlenecks during colonization. Yeh (1988) suggested that a reduction in species-wide
genetic diversity in redcedar was caused by a bottleneck during the last glaciation.
Paleobotanical records suggest that western redcedar experienced a severe population bottleneck
during the last ice age and the species was limited to the extreme southern part of its present day
range (Hebda and Matthewes, 1984). There are two possible scenarios for how a population
bottleneck can lead to a reduction in inbreeding depression. (1) The loss of rare recessive
mutations, some highly deleterious, can occur during a bottleneck allowing a subsequent switch
to inbreeding (Kirkpatrick and Jarne, 2000). (2) During a severe population bottleneck the
reduction to only a few related individuals necessarily leads to mating between relatives and self-
fertilization for reproductive assurance, causing deleterious mutations to be purged. Because both a switch to inbreeding and a reduction in inbreeding depression are so tightly linked it may be impossible to separate these two events in Thuja plicata if they stem from a population
bottleneck.
Thesis Overview
In Chapter two, I will first present estimates of outcrossing rates in six natural
populations of western redcedar based on one polymorphic isozyme locus. In Chapter three, I
will outline the methods used to develop microsatellite markers in western redcedar. Because of 15
the low polymorphism in redcedar isozymes, a more variable marker was required for finer-scale studies. Microsatellites are highly polymorphic, codominant and usually neutral markers, making them well suited for mating system and population genetic studies. In Chapter four, I will address ecological factors affecting outcrossing rates in natural populations of western redcedar. I used microsatellites to obtain estimates of outcrossing for individual trees and position of cones within the crown of trees. I will also describe a new method using bulked
DNA to estimate outcrossing rates. In Chapter five, I will present the results of a controlled pollination study conducted to assess early inbreeding depression and the role of polyembryony in reducing selfing. In general, selfing rates are estimated from germinated seedlings, and in conifers these rates are usually much lower than the actual self-pollination rate. Early inbreeding depression and post-pollination mechanisms, such as embryo competition, can potentially decrease the number of selfed seedlings compared to self-fertilized ovules. In Chapter six, I will outline patterns of genetic diversity over the range of western redcedar and present evidence for separate southern and northern refugia during the last glaciation, suggesting that a species-wide bottleneck predated at least the last glaciation. In Chapter seven I will present an estimate for the rate of somatic mutations at neutral microsatellite loci in western redcedar based on an observed mutation. Chapter eight highlights the main findings of the different studies and ties together genetic diversity, mating system and inbreeding depression in western redcedar. In this final chapter, I discuss the implications of my results and outline further research needed in western redcedar. 16
Chapter 2
The mating system in natural populations of western redcedar
Introduction
Mating systems, inbreeding depression and genetic diversity are inextricably linked
(Uyenoyama et al, 1993; Harder and Barrett, 1996; Holsinger, 1996). Species with higher selfing rates tend to show lower genetic diversity and less inbreeding depression, and the evolution of these quantities involves their close interaction (Charlesworth et al, 1990;
Charlesworth and Charlesworth, 1995). One driving force in their evolution is the lifespan of the organism. Long-lived plants are expected to accumulate a higher genetic load through somatic mutations, consequently leading to the evolution of outcrossing. The large size of trees also increases the chance of self-pollination through geitonogamy, and consequently, the evolution of mechanisms to prevent such accidental selfing (Barrett et al, 1996). Indeed, studies have found that conifers are predominantly outcrossing (5 - 10% selfing; Adams and Birkes, 1991), have high genetic variability (15 - 20% isozyme heterozygosity, Hamrick and Godt, 1996) and high inbreeding depression (mean = 64%, Husband and Schemske, 1996).
However, these conclusions are based upon studies restricted to mainly one family of conifers, the Pinaceae. In the Cupressaceae, genetic diversity has been measured in only a few genera and outcrossing rates have only been estimated in the genus Thuja. In conspicuous exception to other conifers, species of Thuja display high selfing and low gene diversity. Thuja orientalis showed 25% selfing (Xie et al, 1991) and 14% isozyme heterozygosity (Xie et al,
1992), while T. occidentalis showed between 36 and 71% selfing (Perry and Knowles, 1990;
Lamy et al, 1999), and 3 - 13% heterozygosity (Perry et al, 1990; Matthes-Sears et al, 1991;
Lamy et al, 1999). Likewise, western redcedar {Thuja plicata Donn ex D. Don) shows much reduced isozyme variation (4% - 6% heterozygosity), with only one of 21 loci analyzed being appreciably polymorphic (Copes 1981; Yeh, 1988; El-Kassaby et al, 1994). In a seed orchard 17
setting, T. plicata had a selfing rate of 68%, one of the highest measured in a conifer (El Kassaby
etal, 1994).
In conifers, seed orchards are expected to exhibit higher outcrossing rates than their
natural population counterparts (Muona, 1988), or at least be similar to natural populations
(Adams and Birkes, 1991). In natural populations, western redcedar trees tend to occur at lower
density and are larger, leading to more geitonogamy (Farris and Mitton, 1984), and in contrast to
seed orchards where neighbouring trees are unrelated, potential family structure within natural
populations of T. plicata could further inflate natural levels of inbreeding. Knowledge of the
rates of selfing in natural populations, and their variation among populations, will shed light on
this puzzle of the evolution of selfing in predominately outbreeding conifers. In this study I
estimate outcrossing rates in six natural populations of Thuja plicata in southwestern British
Columbia using enzyme electrophoresis. This is the first study to document levels of selfing in
natural populations of western redcedar.
Materials and methods
Sample collections - Cones were collected in the fall of 1996 from two natural
populations of Thuja plicata near Vancouver, British Columbia: the Malcolm Knapp Research
Forest in Maple Ridge and at Mount Seymour (Table 2.1). In 1997, cones were collected from
three populations on southern Vancouver Island (Mount Brenton, Shawnigan, and Reinhart
Lake) and from two populations on the mainland of British Columbia (Mount Seymour and
Enderby) (Table 2.1). Populations consisted of trees of less than 80 yrs except at Enderby,
where trees were over 100 yrs old. In 1996, cones were collected from the lower to mid-crown
of trees, while in 1997 cones were collected from the mid- to upper-crown of trees. I collected
seeds from at least 12 trees from each population, but erratic germination reduced the number of families eventually assayed (Table 2.1). 18
Isozyme analyses - Seeds were extracted from cones and stored at 4°C for up to eight months. They were germinated on wet filter paper at room temperature for approximately 10 days. A few days after germination, seed tissue was ground using the buffer of Mitton et al.
(1979), and the extracts were then stored at -80°C. Only the embryos from the 1996 collections were assayed, while tissues from the 1997 collections were assayed separately as embryos and megagametophytes. The latter tissue is haploid and is genetically identical to the maternal allele passed to the embryo. Assay of megagametophytes allows more accurate inference of maternal genotype in all progeny arrays, and more accurate inference of selfing rate and paternity in arrays descended from heterozygous mothers (Ritland and El-Kassaby, 1985). Maternal genotypes from the 1996 collections were inferred from the progeny arrays.
Following Yeh and O'Malley (1980) and El-Kassaby et al. (1994), I assayed for the
Glucose-6-phosphate dehydrogenase (G6pd) isozyme locus on a morpholine-citrate buffer (pH
8.0). Other isozyme loci were known, from previous studies, to be monomorphic or not sufficiently informative (gene frequency p < 0.05) for mating system estimation (El-Kassaby et al, 1994; Yeh, 1988). In the absence of more markers, and to reduce statistical variance, I increased the number of individuals per family assayed for outcrossing rates. I assayed from about 500 to 700 seed progeny per population, although the 1996 collections yielded fewer germinants for assay (Table 2.1).
Data analysis - Single locus estimates of population outcrossing rates, allele frequencies and correlated matings were obtained using a version of MLTR (Ritland, 1990a) that
incorporated megagametophyte information. Correlated paternity (rp) is defined as the proportion of full sibs among outcrossed sibs (Ritland, 1989). Because of the low number of trees sampled per population, parental fixation indices could not be estimated, and hence were constrained by the estimation program to equal zero. As well, because there was only one 19
marker locus, I could not jointly estimate both components of correlated matings (the correlation of paternity and the correlation of selfing). The latter was assumed to be zero. Errors of estimates were computed with the bootstrap method. Outcrossing and correlation of paternity estimates were considered significantly different from one or zero when the 95th percentile of the bootstrap values did not overlap with these values. The mean outcrossing rate of all populations was obtained by weighting each estimate in proportion to the inverse of its statistical variance. A chi-square heterogeneity test was used to test whether outcrossing rates differed among populations.
Results
The gene frequency of the most common allele at the G6pd locus ranged within a remarkably narrow interval across all populations, from 0.52 to 0.59 (Table 2.1). By contrast, outcrossing rates (f) showed wide variation among populations, ranging from 0.173 to 1.257
(Table 2.1). Estimates of outcrossing were significantly lower than unity in three populations:
Research Forest (t = 0.173), Mount Seymour (t = 0.826) and Enderby (t = 0.771). Estimates of population outcrossing rates differed significantly from each other (%2 = 40, df = 6, p < 0.001).
However, when the Research Forest population was excluded outcrossing rates did not differ from each other (%2 = 5.45, df = 5, p > 0.05). The average weighted outcrossing rate for all populations was 0.715, while the unweighted average was 0.837. Estimates of the correlation of paternity generally did not differ from zero, except for the 1996 Research Forest population, which showed a high value of 0.917 but with a high attached error (it should be noted that estimates of correlated paternity are statistically independent from the estimates of selfing). The weighted average of correlated paternity estimates across populations was 0.025, which did not differ significantly from zero. 20
Discussion
Outcrossing rates - The mean estimate of outcrossing I obtained for western redcedar is
among the lowest in conifers. Estimates of outcrossing in most conifers are above 80% and
exceptions include Larix laricina (mean t = 72.9%, Knowles et al, 1987), Pinus maximinoi (t =
65%, Matheson, 1989), Picea glauca (mean / = 73%, Innes and Ringius, 1990) and Picea rubens
(mean t = 59.5%, Rajora et al, 2000). Low outcrossing rates occur in both Picea chihuahuana
(0% and 15.3%, Ledig et al, 1997) and in Picea martinezii (mean 56%, Ledig et al, 2000), but
these species are restricted to small, extremely isolated populations. The mean outcrossing rate
for Thuja plicata in this study (t = 71.5%) was similar to T. orientalis (mean t = 75%, range: 68 -
81%, Xie et al, 1991) and higher than in T. occidentalis (mean t = 63%, range: 51 - 75%, Perry
and Knowles, 1990; mean t = 30%, range: 24 - 33%, Lamy et al, 1999). One population, the
Research Forest seems to be an outlier in regards to its low selfing estimate and heavily
influences the mean outcrossing rate for all populations. Overall, the estimates of outcrossing
rates I obtained for natural populations of Thuja plicata were higher than I expected. In all populations, except the Research Forest, outcrossing rates were higher than in the seed orchard
study (t = 32%; El-Kassaby et al, 1994).
Factors affecting mating systems - Our results, together with the high selfing rate found
in the seed orchard (El Kassaby et al, 1994), indicate that the mating system in western redcedar
is quite labile, with marked among-population variation, and probably marked among-tree
variation. A possible reason for the lower outcrossing rate in the western redcedar seed orchard
is assortative mating, caused by asynchrony of receptivity to pollen among trees from different
geographical locations in the orchard. In a Douglas-fir seed orchard El-Kassaby et al (1988)
found that trees that were receptive earlier or later had higher selfing than the remaining trees. 21
There are numerous factors that may affect the selfing rates of western redcedar, and of conifers in general, described as follows.
(1) Clonal structure - Western redcedar probably exhibits clonal structure, because of extensive vegetative reproduction giving rise to groups of genetically identical ramets, which can increase the rate of inbreeding.
(2) Family structure - Localized family structure due to limited dispersal can contribute to biparental inbreeding. More than one polymorphic locus is needed to differentiate between self- fertilization and mating between relatives.
(3) Tree size - In the populations with the higher outcrossing estimates (Shawnigan, Mount
Brenton, Reinhart), I collected seeds from young trees with larger, mature trees nearby. In contrast, significant inbreeding was found in Enderby where seeds were collected from older trees only. A higher outcrossing rate in smaller trees is expected because they produce less pollen and therefore their pollen constitutes a smaller proportion of the pollen cloud and increases the chance of being fertilized by outcross pollen.
(4) Crown position - Outcrossing rates can also vary among different heights within a tree.
Lower outcrossing rates have been found in lower crowns compared to upper crowns in Douglas- fir (Pseudostuga menziessii) and Sitka spruce (Picea sitchensis)(Ovm and Adams, 1986; El-
Kassaby et al, 1986; Chaisurisri et al, 1994). In this study, cones were collected from the upper-crown of trees in most populations but seeds were collected from lower branches in the population with the lowest outcrossing rate, the Research Forest. However, seeds collected in the same way in the 1996 collection from Mt. Seymour did not also show lower outcrossing.
(5) Inbreeding depression - High inbreeding depression at the seed stage (mean 5 = 0.58,
Husband and Schemske, 1996) can lead to the high rates of outcrossing observed in conifers. As seedlings are normally used to infer outcrossing rates, the selfed seeds that die are missed in the outcrossing rate estimate. However, T. plicata seems to lack early depression at the seed stage 22
(Owens et al, 1990; Husband and Schemske, 1996), so my estimates of outcrossing are not as biased by this early-acting inbreeding depression as in other conifer species.
Correlation of paternity - Very few mating system studies in conifers report correlation
of paternity estimates. In Pinus washoensis (Mitton et al, 1997) and Tsuga mertensiana (Ally et
al, 2000), no significant correlation of paternity was found. In two species, Picea martinezii (rp
= 0.389; Ledig et al, 2000) and Abies borisii (rp = 0.990; Fady and Westfall, 1997), the high
correlation of paternity estimates were attributed to the low number of reproductive individuals
in the population. In Larix occidentalis, correlation of paternity estimates were significant in
high-density populations (rp = 0.062 and 0.104) but not significant in low-density populations (rp
= 0.001 and 0.02; El-Kassaby and Jaquish, 1996), probably because high tree density limited pollen movement. Overall my results showed no correlation of paternity in redcedar (mean
weighted rp = 0.025) indicating that the outcrossed seeds within a progeny array were fertilized by several different pollen parents. The low outcrossing rate and high correlation of paternity in the Research Forest population suggest that trees may have been disproportionately exposed to their own pollen in this population.
In this study I have found significant amounts of inbreeding in natural populations of
western redcedar and variation in outcrossing rates among populations. I have set up a framework for further studies on the mating system within natural populations with more polymorphic and informative genetic markers, microsatellites (O'Connell and Ritland, 2000;
Chapter 3). CO OO T-H in co Os in © 00 CN oo o so Os CN 00 CN >n CN © © © © CN o o © © © © © © © ©' © r- 00 co Os in T-H CN oo © © •T-H in SO CN Os o © © © T-H © ^H © © © © © © ©• © d" © 1 V
in CN T-H in r- CO in m Os >n r- m T-H CO m SO o CN © T-« T^ T-H CO © © © © © © © © © • © ©
CO 00 SO 00 SO T-H r- in r- CN so r- in CO • i—i Os 00 Os Os r- CN 00 t— © ©' © ©' ©' ©' T-H © ©'
' N co in CO SO 00 CO 00 SO o T-H © © © O a s * © o T3 © © © © © © O, 3 Q. o SO co oo oo CN Os co SO CO r- 00 o >n U0 uo co 00 00 IT) Os CO SO TT co © in T-H in CN r- m in Os >n CN 00 00 Os Os o T-H © o © © © oo co in i in © Os 03 1 1 1 H in T—H i T—< T-H IT) 50-8 0 © r- >n T-H A a in r- CS > © • o o © © o © in o © © 00 CN •n in SO SO so r- u 43 4) -a B 3 in r- r- CN in T-H r- in in >n © •* © OH 5) ^ 0 o 0 o o o O a CN CN CN CO 00 _B o CN CN CN CN CN r-H CN T-H T-H T-H T-H T^ T-H a ^ > -1a) '5b o CN CN CO 00 © in Q CN CN CN >n in CO o o 0 o o o O Os OS Os 00 00 © 00 on •si• >n 43 -H u 3 3 3 B o O 8 o B T3 a >> a r-1 o PH JD T3 S3 43 on ovin m t SP 43 "8 .SP u B 4—s» a 43 C3 OX) S0 3 •*—>3 B •e _3 ©s O OS O o o as I OS 'S 43 B OH u > CM1 s s s * 24 Chapter 3 Characterization of microsatellite loci in western redcedar Introduction Conifers are among the most genetically diverse plants (Hamrick and Godt, 1996) and are predominantly outcrossed (Barrett and Eckert, 1990). In contrast, western redcedar (Thuja plicata Donn ex D. Don Cupressaceae) has shown low genetic diversity based on leaf oil terpenes (von Rudloff and Lapp 1979), isozyme loci (El Kassaby et al, 1994) and restriction fragment length polymorphisms (Glaubitz et ah, 2000). Population outcrossing rates for western redcedar based on one isozyme locus indicated a mixed mating strategy in this species (El- Kassaby et ah, 1994; O'Connell et al, 2001; Chapter 2). Thuja plicata is a widespread conifer found along the west coast of North America from southern Alaska to northern California, and in the interior from east-central British Columbia into northern Idaho (Minore 1990). The coastal and interior populations are geographically isolated from each other and may be genetically differentiated. Estimating outcrossing rates in plants usually requires several polymorphic loci, however. In species with low genetic diversity the lack of isozyme polymorphism prevents us from obtaining accurate estimates of outcrossing rates with this marker. Because microsatellites are highly polymorphic, codominant and usually neutral markers they are well suited for mating system studies. I designed microsatellites for T. plicata to study its population genetic structure and mating system. Materials and Methods Clone development - Microsatellite markers were isolated from redcedar genomic DNA using modifications of biotin-enrichment strategies of Kijas and Fowler (1994). Genomic DNA was digested with Hae III, and individual fragments were ligated to double stranded 25 oligonucleotide adapters (M28, M29) on their 5' and 3' ends, respectively. Adapted fragments were then denatured, hybridized with 5' biotin labeled (TG)12 and enriched by selection with magnetic streptavidin affinity supports (Dynal M-280). Biotin selected genomic fragments were then amplified using primer M30, and the resulting mixture was cut with EcoRI and ligated into standard cloning vectors (pGEM3Z+, Promega) for propagation in bacteria. Individual 32 microsatellites containing clones were isolated by colony hybridization with P -labeled (AC)12 and picked into glycerol cultures for long term storage and isolation. I sequenced 96 clones directly from glycerol stocks using SequiTherm EXCEL™ II Long-Read DNA Sequencing Kits-LC (Epicentre Technologies) on a LI-COR 4200 sequencer (Lincoln, Nebraska). From these, I chose 35 clones to design microsatellite primer sets. In each primer pair, one of the primers was tailed (Table 3.1; Oetting et al, 1995). Screening for polymorphisms - Total DNA from T. plicata foliage was isolated using a modified CTAB method (Doyle and Doyle, 1987; Appendix IV). To test for microsatellite polymorphism I screened individuals from one coastal population (southwestern BC, N = 22) and two interior populations (southeastearn BC, A^= 11; and northern Idaho, N = 11). Previously isolated DNA samples for the interior populations were part of another study (Glaubitz et al, 2000). Polymerase chain reactions (PCR) amplifications were performed using 10 u\ total reaction volumes with lx Taq buffer (lOmM Tris, 1.5 mM MgCl2, 50 mM KC1, pH 8.3; Roche), 1 pmol dNTP, 0.5 pmol each of forward and reverse primers, 0.5 pmol M13 IRD-labeled primer, lUnit Taq DNA Polymerase (Roche), and between 10-30 ng of genomic DNA template. Samples were amplified on a PTC-100 thermocycler (MJ Research) denaturing at 95 °C for 5 min, followed by 33 cycles of 95 °C for 45 s, annealing temperature (Table 3.1) for 45 s, 72 °C for 45 s and ending with one cycle of 72 °C for 5 min. Following amplification, 3 u\ of loading 26 dye (100% formamide, lmg/ml pararosaniline basic red 9) was added to each reaction. For final screening the microsatellites were detected on a LI-COR 4200 sequencer with a 7% poly aery lamide (Long Ranger™) gels. Results and Discussion Of the 35 primer sets, 12 amplified interpretable polymorphic loci. One of these showed two loci for the primer pair (TP12a & b; Table 3.1). Six loci were composed of simple dinucleotide repeats, two of compound repeats, and three had interrupted repeats. One of the loci, TP6, included a hexanucleotide repeat. Observed and expected heterozygosities were higher for a greater number of loci in the coastal population than in the interior population. There was a significant deficiency in heterozygotes at three loci (TP4, TP10 and TP12a) in the coastal population indicating possible null alleles. Because redcedar is known to self-fertilize in natural populations, a heterozygote deficiency can also be due to inbreeding. The number of alleles per locus ranged from 3 to 36 for a total of 189 alleles for the 13 loci. Together these loci have enough variability to conduct a detailed study of the population genetic structure of T. plicata. Loci with high number of alleles will be particularly useful for mating system studies in this species. The 12 sequences with interpretable microsatellites have been deposited in Genbank (Appendix V). r- co o m r- m r- o in m so Os r- oo so d o o ro in in o o r- CO in •§ SI m o r- r- in OO r- a d d d d d o CN CN CN os Os CN CN CN CN CN CN CN r- CN so CO o CN CO CO in CO r- SO Os OO 00 r- oo oo oo d d in CO r- o Os I/O o Os CN CS no Os o r- r- Os SO m r- o d d d d d V d d CN CN o CN CN o CN CN CN CN CN CN SO CN SO CO SO CN Os CO SO Os Os CO o 00 r- f- CO CO CN CN CN CN CN CN I I I i co SO CO SO >n I m Os 00 in >n SO SO CN CN 00 CN CN CN CN a c« CS I < o Os A u S i o O a o < u Ql < H a a < u u u 04 ' u o CN in in in oo m CN o SO m in in in >n >n >n o # u < O u * < * a H * a * H < H a a < a < < u a o H u u < < a H < u o < < < < H u H U a < < u a O rj U H < < < U < < u H < H < U < a a < u a < U H < U < HU H H < < < < < a a H < < U < H 3 < o CN CO >n so r- 00 o CM OH _1 H H H oo oo Tt NO co CN oo CO oo 00 CN CO © © © ON © r- r- © in CN Si ^ ON CO r- 00 CN © CN ON CN CN CN CN o OD >> N ON © ON o IT) CO © Ii ON NO NO ON CJ © © © d ej J3 >n ON in CO r- CN ea ON in in CO o 5 CO NO X) © o U d s c CN ON 3 cj CN G O T3 ^" -1—1 if O NO CN \— is 3 'S NO CN © r- CN CJ CO CN CN i i Tt 00 CO ON in Tt NO © NO cj CN on CN C/3 O 00 • "3 cj & £ o C o U U c h2 o O o 1 E-T •a ea CJ o| E-i < •o U U B 3 < CJ u a. "S rv 00 in >n to CJ in o ea NO NO NO ca =E ca cj o o a o CJ £• cj 'E8 •£ CT3 1 CJ N u H O T3 a < H < o a a o U CJ < a u H cj H < JS 3 H a H U U CJ o H > H H U a O -a ca < < < o O CJ CJ <2 H a < o > 8 g < o H u O H H < < a U H 1 o U u H cj s E-i a cj o < .O CJ u OH a b a H T3 X u u U CJ cj H < H o H OH W) < < < tH < a < a a S 6 x>CJ < a u c CJ •c u a u H "53 OH < ca u o u o 3 JO T3 CN CN ON > f PH bZ CU •s H H H H PH * , 29 Chapter 4 Fine-scale estimation of outcrossing in western redcedar with microsatellite assay of bulked DNA Introduction Western redcedar (Thuja plicata Don ex D. Donn, Cupressaceae) shows quite high selfing rates for a conifer, with estimates of outcrossing in natural populations ranging from 17- 100% (weighted mean = 71.5%) (O'Connell et al, 2001; Chapter 2). The outcrossing rate for most conifer species is above 80% (52 species, mean 83.5%; Chapter 1). While western redcedar produces abundant viable selfed seeds (Chapter 5), inbred trees produced by self- fertilization have shown significant reduction in growth rate (ca. 10%) compared to non-inbred trees (J. Russell, unpublished data). Because seeds for reforestation are normally collected from natural populations, knowledge of variation in selfing rates within a tree is important for the collection of seeds with the highest expected outcrossing rate. In addition, knowledge of fine- scale variation of outcrossing rate will enable deeper understanding of the evolution of factors mediating selfing rates (Barrett and Eckert, 1990). The mating system of conifers is dynamic in space and time, mainly being affected by variation of self-pollen availability (Mitton, 1992). Pre-pollination factors that can affect inbreeding rates in conifers include population density (Farris and Mitton, 1984), crown position (Chaisurisri et al, 1994; El-Kassaby et al., 1986; Omi and Adams, 1986), family structure and tree size (Mitton, 1992). In redcedar both male and female cones are distributed throughout the crown of trees and the lower branches often reach the ground. Higher selfing is expected in lower branches or in cones closest to the trunk because they are more likely to receive self-pollen from branches higher up in the tree. Likewise, higher branches and branch tips are expected to receive more unrelated pollen and show less selfing. In a previous study of western redcedar, outcrossing estimates showed that lower branches have higher levels of self-pollination than 30 higher branches (K. Ritland, unpublished data). However, these were in seed orchard populations, which may be unrepresentative of natural populations. In another study, differences for outcrossing among natural populations suggested that populations with larger trees showed lower outcrossing rates (O'Connell et al, 2001; Chapter 2). Population outcrossing rates are usually estimated using isozyme markers, as in O'Connell et al. (2001) and Chapter 2, but in western redcedar, only one isozyme locus (glucose- 6-phosphate dehydrogenase) of 21 surveyed shows sufficient polymorphism for estimation of outcrossing rate (Copes, 1981; Yeh, 1988; El-Kassaby et al, 1994). More loci are needed to study finer-scale variation of outcrossing rates within a population. Several highly variable microsatellite loci have been developed for western redcedar, which should allow us to conduct detailed studies of variation of selfing rate (O'Connell and Ritland, 2000; Chapter 3). Microsatellites can indeed be used to estimate outcrossing at different crown positions {Pinus densiflora, Pinacaeae; Lian et al, 2001), to discriminate between bi-parental inbreeding and self- fertilization (Caryocar brasiliense, Caryocaraceae; Collevatti et al, 2001), and to study the variation in selfing rates between immature and mature fruits (Shorea leprosula, Dipterocarpaceae; Nagamitsu, 2001). In the herbaceous plant, Zostera marina (Zosteraceae), outcrossing rates and relatedness among siblings were estimated using microsatellites (Reusch, 2000, 2001). However, the disadvantage with microsatellites is that they are more expensive and labour intensive, compared to the less-informative isozymes. The cost of assay leads to a reduction in sample size. For example, in other microsatellite studies, seeds were sampled from only one maternal tree in Pinus densiflora (Lian et al, 2001), five trees in Shorea leprosula (Nagamitsu, 2001) and outcrossing rates were estimated from a single progeny per family in Zostera marina (Reusch, 2000, 2001). By bulking offspring collected from the same location 31 within a tree, the number of DNA extractions and samples genotyped can be significantly reduced, provided that alleles can still be detected in bulked samples. In this study I test for fine-scale variation of outcrossing rates within natural populations of Thuja plicata, at the level of individual trees and within individual trees. To increase experimental efficiency, I use a new estimation method based upon the bulking of several individual progeny into one sample. This method greatly increases the power to detect fine-scale variation, as more individuals can effectively be included for the same number of genetic assays. This method is then used to test for fine-scale differences of outcrossing rates within populations of western redcedar, in relation to tree size and position of cones within trees. Materials and Method Bulking tests - To evaluate the feasibility of bulking DNA (extracting DNA tissues from several individuals simultaneously), I performed three tests to assess whether all alleles could be detected, using samples bulked either before or after DNA extraction. First, I bulked equal proportions of DNA separately extracted from three individuals with known genotypes to test whether all alleles could be detected. Second, I bulked DNA from two individuals in 5:1 and 1:5 DNA ratios to test whether alleles occurring at a low frequency could be detected. Third, I screened samples of one, three or ten seedlings bulked before DNA extraction and in this case the individual seedling genotype was not known but the maternal genotypes were obtained from separately screened haploid megagametophytes. I screened all individuals at four easily interpretable and robust microsatellite loci (TP1, TP3, TP9, TP11, Chapter 3). PCR amplification and detection of alleles on a LI-COR 4200 (Lincoln, Nebraska) were carried out as described in O'Connell and Ritland (2000) and Chapter 3. To score alleles, I used a program that gives the intensity of each band (Odyssey ver. 1.0.55, LI-COR Inc., Lincoln, Nebraska). The band intensity is equal to the sum of the intensity values 32 for all pixels in an area covered by a band (integrated intensity). To test whether alleles were missed in bulked DNA samples compared to unbulked samples, paired t-tests were used. Statistical tests were performed using JMP version 3.2.1 (SAS Institute, 1997). Sample collections - In the autumn of 1999, mature cones were collected from four southwestern British Columbia populations: three on eastern Vancouver Island (BC12, BC14 and BC15) and one on the mainland near Pemberton (BC13; see Chapter 6). Sampled trees ranged in height from 4.8 to 36.8 m, and were representative of the height of reproductive individuals in the populations. Cones were collected from up to three different crown heights within a tree: the top of each tree, midway up the tree and from the lowermost branches of the tree. At each height, cones were collected from two positions: from the tip of the outer branches and from the hanging inner branches closer to the trunk. In the three island populations, cones were collected from up to six positions on each tree while in the Pemberton population, seeds were collected from only two positions (1 and 5) (Figure 4.1). Seeds were mechanically extracted from the cones and stored at 4°C. Seeds were germinated in petri dishes on moist filter paper at room temperature (O'Connell et al, 2001). A few days after germination the haploid megagametophytes and diploid seedlings were separated and placed in 1.5 mL microtubes. Bulks of ten megagametophytes were used to obtain maternal genotypes and bulks of three seedlings to obtain estimates of outcrossing. Samples were kept at -20°C until DNA extraction. Each sample was ground and DNA extractions were carried out using a modified CTAB method in the microtubes (Doyle and Doyle, 1987; Appendix IV). 33 Fig. 4.1 Diagram of six cone collection positions in a Thuja plicata tree: (1) top, outer branches (2) top, inner branches (3) mid, outer branches (4) mid, inner branches (5) lower, outer branches and (6) lower, inner branches. DNA assay of bulks - To test for differences in outcrossing among crown positions, two collections of three bulked seedlings each were screened from each position. Seedlings were screened at four loci: TP1, TP3, TP9 and TP11. Megagametophytes were also scored at three additional loci as part of another study (Chapter 6). The total number of seedlings sampled per tree ranged between 12 (two positions) and 36 (six positions). To ensure that bands were uniformly scored, I used an allelic ladder composed of two to three individuals with alleles of known sizes and spanning the range of allele sizes at a locus. The intensity and size of microsatellites were scored using the Odyssey software. This extra caution in scoring bands was 34 needed, as many alleles are possible in bulk samples. For example, in a bulk of three seedlings, up to four alleles can exist for a homozygous mother, and up to five alleles for a heterozygous mother. Estimation of outcrossing from bulk samples - Probabilities of progeny, conditioned upon maternal genotype, are the basic ingredients for the estimation of mating systems. These are functions of the population allele frequencies and the outcrossing rate. Table 4.1 gives these probabilities for cases of single progeny, two progeny and three progeny. The probabilities of single progeny are used in the classic method of estimating outcrossing, while the two- and three-progeny probabilities represent the cases where two and three individuals are bulked, respectively. 35 Table 4.1. Probabilities of band patterns observed for a single progeny, and for bulked progenies of sizes 2 and 3, conditioned upon maternal genotype (homozygous A,A,-or heterozygous A,Ay). Note: for outcrossing rate t = 1 - s, and for gene frequency pt, the following abbreviations for formula are used: (1) a,• = s + pf, (2) bt = pf, (3) c, = s/4 + pjt/2, (4) dt = ptt/2. Bands Prob|A,A, Bands Prob|A,A, 1 progeny (unbulked) 1 ai c, ij bj C,+Cj i,k or j,k 2 progeny (bulked) l •J 2apj+bf >] c,+2cicj+c/ 2bjbk i,j,k or i,k or j,k 2Cjdk+2Cjdk i,j,k,l or i,k,l or j,k,l 2dkdl 3 progeny (bulked) i a? i 2 i,j 'iafbj+'bapj+b- i,j c/+3c, cy+3c,c/+c/ 2 2 2 2 2 i,j,k 6aibpk+?>bfbk+3bjbk i,j,k or i,k or j,k 6cfjdk+3c1 dk+3cJ dk+3cjdk +3cjdk +dk 2 2 ij, k, I 6bjbkb, ij, k, I or i, k, I or j,k, I 6cidkdl+6Cjdkdl+3dk dl+3dkd ij, k, I, m or i, k, I, m or j, k, I, m 6dkdflm The procedure for estimating outcrossing then follows the usual procedures, as for example, described in Ritland (1983). A computer program was written in FORTRAN 95 by K. Ritland for estimating outcrossing rate for the case of bulks of size three. Maternal parentage was assumed known, and the bootstrap method was used to ascertain errors of estimates. Also, another program was written to simulate data, and hence to check the accuracy of the estimation program. Estimates indeed were obtained within the statistical range of the true values of outcrossing used to generate the simulated data. Ritland (2002) presented a model for estimating gene frequencies and heterozygosities using bulked samples. In that paper, a general probability was given using a "mask" function, 36 but this notation is much more difficult with these mating system probabilities. In that paper, it was found that with larger pool sizes, the higher non-linearity of the probability model resulted in greater estimation bias. Likewise, the simulation program did find bias but the bias was not more than 5% of the true value of outcrossing rate, for the bulk size of three. Results Allele detection - In samples of two and three seedlings that were bulked after DNA extraction all alleles were detected at each of the four loci in both the 1:5 and the 1:1:1 DNA ratios (results not shown). Microsatellite alleles typically show additional non-allelelic bands of lower intensity and smaller size on gels. These stutter bands are the result of slippage during PCR amplifications. Tests performed with individuals of known genotypes showed that stutter bands were almost exactly half the intensity of the previous band and the intensity of overlapping bands was additive. Using this information, alleles can be disentangled from stutter bands and other alleles in bulked samples (Fig. 4.2). The tests also showed that, for bulked samples, the number of copies of a particular allele could not be accurately inferred. This is probably because of allele competition during PCR reactions. Bands from larger size alleles are usually less intense than for shorter alleles (pers. obs.) 37 Locus TP9 Band intensity profile Fig. 4.2 Band intensity profile (right) of three bulked individuals at locus TP9 from the lane 6 on the microsatellite gel (left). The four detected alleles are indicated by black arrows on the band intensity profile. M, 10 bulked megagametophytes from the maternal plant. S, three bulked seedlings. Bulking test - The total number of alleles detected for the same number of seedlings was greater when samples were not bulked (Table 4.2). Maternal alleles and common alleles in the population were detected in both bulked and non-bulked samples. However low frequency alleles were probably missed in bulked samples. Alleles at loci spanning a larger range of allele sizes were more likely to be detected because there was less band overlap. Bulks of three seedlings were chosen for outcrossing rate estimation because individual alleles were easily identified, yet bulking still provided the advantage of reducing the number of samples of DNA to extract and score. 00 co 1) 43 X X ro *t *t in es CN ro on ro ro _3 T3 "o 4) u "3 co SO ~G 42 X CN CN CN CO CO CN 4) CN o 43 >n o CO SO 0 CO CN CO CO Tt CO CN X CN V T3 a. o 60 CO _B CO CO CO CO CO CN T3 -a X CN 4) xi 4) B 1) 4) O =3 cd •a 60 PH _B 4) c x 4= In Xi 4) 4) 43 4) 3 < B X Tt W VO t~ t» os Os CN X 43 Q O 03 4) 4) M 4) ~3 CO SO 1) 1 1 X CO lO " O H 43 ^ O T^ .2. PH 'a PH H 60 S X> X CN o i/i so Tr Ul 42 _B X* XI 4) 4) 4) o -4-t CO SO -83 Genetic diversity - Microsatellites were scored in a total of 2019 seedlings from 80 families in four populations (20 families/pop). Expected heterozygosity and inbreeding coefficients for each population based on seven loci were obtained from the megagametophytes (Table 4.3; Chapter 6). Three of the populations had inbreeding coefficients that were significantly greater than zero. The four microsatellite loci used to score seedlings showed a large amount of polymorphism within sampled populations. The number of maternal alleles per population ranged from five to seven at locus TP11, and from 14 to 16 at locus TP9. At each locus almost twice the number of alleles were detected in the seedlings compared to the maternal plants. Outcrossing rates - Outcrossing rates did not significantly differ among branch heights over all the populations (Table 4.4). In the three populations where cones were collected separately from inner and outer crowns, outcrossing rates were higher in inner branches compared to outer branches (6 of 9 positions). However, the differences in outcrossing rates between inner vs outer branches were not statistically significant (paired t - test: 1.57, P = 0.077). Overall tree outcrossing rates decreased significantly with tree height (^=0.15; N =73; P = 0.0006; Fig. 4.3). An analysis of covariance showed that overall the decrease in outcrossing with an increase in tree height was highly significant (F = 12.8; df= 1; P = 0.0007). But there was no difference among populations in the mean tree outcrossing rate (F = 0.35; df = 3; P = 0.79) or for the slope of outcrossing rate on tree height (F = 0.19, df= 3, P = 0.90; Fig. 4.3). The mean outcrossing rate over all trees in a population ranged from 66% to 78% (Table 4.5). o 3 Tt Tt O ^ 00 OH CO CO CO CN _B CM c E-I cj cj 0\ CM > CJ ^ >n oo r- CJ H CaO ON o CM •a H CO CJ r- Tt Tt CO CM o CO ai H CN CM H O O CM OH m CN CN 00 H N o o IM O o CJ ^H O CJ o J3 60 d ON X) a o Tt CN CO CJ V =3 ^ CO Tt CO X) 1—I NO o, cj CN NO o cj cj >n CN a, ©" 00 x A CJ i- * O Tt Tt Tf NO X> o CM i—1 ^M »™M CJ o C3 H r—t On T3 £? cj 0) -M O .-3 •*-» ON 3 3 CJ CJ 111 CM r- m NO NO CJ oo CO -w H CJ CO 00 c/3 T3 e _B o JZ CJ xi cd o ON ON CN CJ 6 ^M i—i CJ a <+H o o O > lH Ui CJ CJ CJ I/O B ON NO 00 ON O 1 13 -a 3 B CJ CO G * * * cd o CO ON NO CN -o -*-» in Tt B X) ^oM o o O °o d d d d to co oo "+H cj in ^M CD i- .fi O i-H CN Tt 3 00 00 O C3 cj oo d d d d CJ s I oo '•3 o o o o X) CJ O CJ cj CN CN CN CN u- 00 .O ^ B r IC3 Xa) o Tt CO ' V ^M CO a U CJ O CJ o a CN U o 3 B cj 3 m o .B OH 6 o o O a o a H CJ CJ Tt OH o o -4—» CM cj CM co 3 O hH o 3O 2 3 ca C3 o CJ CJ o H U CM c 42 1.0 0.8 L 0.6 L Outcrossing rate (t) • 0.4 o 0.2 •—Coombs G Yellow Point ••• • Paldi ^Pemberton 0.0 0 10 15 20 25 30 35 40 Tree height (m) Fig. 4.3 Individual tree outcrossing rate estimates regressed on tree height in four populations of Thuja plicata. N = 13. 43 Table 4.5 Mean tree heights and individual tree outcrossing rates (t) in four populations of Thuja plicata. Population Mean tree height (range) t (range) SE Coombs 21.7(14.0-35.5) 0.741 (0.27- 1.00) 0.040 Pemberton 20.2 (4.8 - 36.8) 0.778 (0.29 - 1.00) 0.040 Yellow Point 20.2(12.1 -34.5) 0.662(0.22-0.98) 0.046 Paldi 18.8 (8.3-32.8) 0,744 (Q.48 - 1.00) 0.033 Discussion Variation in outcrossing rates - Selfing rates in plants are usually estimated using seedlings, and are lower than the true self-pollination rates because of selection against inbred embryos. Redcedar shows high self-fertility and the correlation between selfed seeds and self- pollination is closer than for other conifers (Chapter 5). No significant difference in outcrossing was detected among crown positions. Furthermore, all trends observed were in the opposite direction of what was expected. Outcrossing rates were higher in the lower branches vs the higher branches, and higher in inner vs outer branches. Unlike these natural populations, trees in a previous study in a western redcedar seed orchard showed a significant decrease in selfing in higher branches compared to lower branches (K. Ritland, unpublished data). Higher outcrossing in upper crowns vs lower crowns was also found in Sitka spruce (Picea sitchensis; Chaisurisri et al., 1994) and Douglas-fir seed orchards (Pseudotsuga menziesii; El-Kassaby et al., 1986; Omi and Adams, 1986). Factors in seed orchards that differ from natural populations and that could enhance outcrossing include higher tree density, top pruning and decreased family structure. These factors could also contribute to the difference in outcrossing among crown heights observed the seed orchard. The 44 variation in the number of inbred seedlings among different positions within a tree is due mainly to variation in the amount of self-pollen received. In Thuja plicata, outcrossing rates decreased with increasing tree height in all four populations. Individual tree outcrossing rates varied widely in all populations (Table 4.5). Larger trees were probably more likely to self-fertilize than shorter trees because their pollen makes up a larger proportion of the pollen cloud. The tallest redcedar trees in every stand towered above other trees. All four populations were similar in terms of tree height structure (Table 4.5). Phenological differences that occur with tree size can also potentially affect outcrossing. In Douglas-fir, shorter trees have a higher proportion of male cones, and larger trees have more female cones so that outcrossing rate is expected to increase with tree height (Mitton, 1992). This is opposite to the trend observed in redcedar. No data are available on whether there is a change in cone sex-ratio with tree size in redcedar. Alternatively, if selfing is prevented by a self-incompatibility mechanism, a breakdown in this mechanism with tree age could explain the increase in selfing in larger trees. Although there is no strong evidence for this scenario it would be an interesting question to pursue. Population outcrossing rates - Unlike a previous study conducted with isozymes (O'Connell et al, 2001; Chapter 2), there was no difference in outcrossing rates among populations. In the current study, outcrossing rates were not significantly different among populations and occurred within a narrow range (t = 66.2 to 77.8%). In contrast to the isozyme study, sampled trees in this study were more homogeneous in size across all populations. If selfing rates are constant over time the level of inbreeding should be reflected in the inbreeding coefficient. There was no relationship between mean population outcrossing rates and inbreeding coefficients or genetic diversity. 45 Bulking samples - Estimates of outcrossing obtained using bulked DNA samples should be lower than when using un-bulked samples for two reasons. First, alleles are more likely to be missed in bulked samples due to allele competition during PCR and overlapping stutter bands on gels. Second, simulations showed a 5% downward bias in outcrossing estimates of bulked samples. Correspondingly, in the redcedar seed orchard study average outcrossing rates estimated using microsatellites with non-bulked samples were about 5% higher than those observed in this study (K. Ritland, unpublished data). Likewise, in the isozyme study outcrossing estimates were above 77% for all populations but one (5/6) (O'Connell et al., 2001 and Chapter 2). At least, the comparison of trends in ecological factors affecting outcrossing rates are little affected by systematic statistical bias. 46 Chapter 5 Polyembryony and early inbreeding depression in a self-fertile conifer, Thuja plicata (Cupressaceae) Introduction Inbreeding depression, expressed by self-fertilized progeny through reduced survival and reproduction, is a major driving force in the evolution of plant mating systems (Lande and Schemske, 1985). Plants have evolved several pre- and post-pollination mechanisms of inbreeding avoidance. In general, conifers show very low seed set after self-pollination (reviewed in Husband and Schemske, 1986, and Kormut'ak and Lindgren, 1996), and traits favoring cross-fertilization are usually present. These include asynchrony of pollen shedding and ovule receptivity (e.g. Eriksson and Adams, 1990) and physical separation of male cones from female cones within a tree (e.g. Park and Fowler, 1984; Omi and Adams, 1986). Unlike most conifers, western redcedar (Thuja plicata Don ex D. Donn, Cupressaceae) has shown significant amounts of inbreeding in both orchards and natural populations (El- Kassaby et al, 1994; O'Connell et al, 2001; K. Ritland, unpublished data; C. Newton, B.C. Research Inc., unpublished data). Redcedar lacks the obvious inbreeding avoidance traits. Pollen shed and ovule receptivity overlap, and male and female cones are intermingled within a tree (Owens et al, 1990; El-Kassaby, 1999). Thuja plicata also shows high self-fertility, setting abundant seed after self-pollination (Owens et al, 1990). Because pure stands of western redcedar are rare and trees are often isolated from conspecifics (Minore, 1983), the species has probably had to adapt to the low availability of unrelated pollen and self-fertilize. A possible mechanism that can reduce the effects of inbreeding in conifers is polyembryony, the occurrence of multiple embryos within an ovule. In conifers, several genetically identical archegonia within the same ovule can potentially be sired by pollen grains from different parents. Polyembryony can potentially "rescue" ovules with an inviable embryo. The abortion of an embryo does not necessarily cause the abortion of the seed if at least one 47 embryo within the same ovule is viable. The presence of more than one viable embryo per ovule also offers the opportunity for competition between embryos and should favor those that are non- inbred. This implies that polyembryony might increase ovule success rates when embryos have low viability and/or be a mechanism to eliminate self-pollinated embryos (Sorensen, 1982; Latta, 1995). Without pollen, redcedar cones start to develop but soon stop growing and remain on the branches as small, undeveloped cones; pollen is required for the initiation of megagametophyte development. Unlike other conifers, redcedar cones will develop with only one fertilized seed, and fertilized but aborted seeds are externally indistinguishable from viable seeds (Owens et al, 1990). Thuja plicata has archegonial polyembryony with seven to nine archegonia per ovule. On average, two to three archegonia per ovule are fertilized by different pollen grains. Eventually one embryo becomes more mitotically active, and the other embryos in the ovule are aborted (Owens and Molder, 1980). In western redcedar self-fertilization can offer reproductive assurance when conspecific pollen is rare, and polyembryony could favor outcrossing when a mixture of outcross and self-pollen is received. Polyembryony is often suggested as a mechanism mitigating the effects of self- fertilization and is incorporated in theoretical models (Park and Fowler, 1984; Crook and Friedman, 1992; Nakamura and Wheeler, 1992; Savolainen et al, 1992; Karkkainen and Savolainen, 1993; Morgante et al, 1993). However, there is a lack of empirical data testing this hypothesis. In this study, I used seed set following self- and cross-pollinations to obtain viability estimates of selfed vs outcrossed embryos in western redcedar. From these estimates, I calculated the proportion of full seeds and selfed seeds expected with and without polyembryony following Sorensen (1982). I then used mixtures of self and outcross pollen to test the null hypotheses that there is no difference in seed set and proportion of selfed seeds from that expected in the absence of polyembryony. An increase in the proportion of viable seeds 48 (compared to expectations without polyembryony) suggests that non-inbred viable embryos rescue ovules that also contain inviable inbred embryos. Similarly, a decrease in the proportion of selfed seeds, indicates that outcrossed embryos are out-competing selfed embryos within the same ovule. Materials and Methods Pollinations -1 conducted a controlled pollination study at the Mount Newton seed orchard in Saanichton, British Columbia during the winter of 1999. I chose four trees with a large number of both male and female cones. Each tree was genotyped at the Glucoses- phosphate dehydrogenase (G6pd) isozyme locus. Other isozyme loci in this seed orchard were either monomorphic or showed very low allelic diversity (El-Kassaby et al, 1994). Two of these trees (181 & 395) were homozygous for the slow allele "a" at this locus, one tree (432) was homozygous for the fast allele "A" and one tree (431) was heterozygous (Table 5.1). On each tree 14 branches containing a few hundred female cones were covered in paper bags with plastic windows in mid-February, before they were receptive to pollen. Before bagging the branches, all male strobili were removed. Other branches, with pollen cones, were collected and placed on drying racks at 15-17°C and 60-70% relative humidity for two to three days until they released pollen. The collected pollen was then stored at 4°C in plastic vials for up to three days. Mixtures of pollen from two trees were measured in a graduated cylinder, placed in small plastic squeeze bottle with an attached nozzle, and mixed by shaking. 49 Table 5.1 Experimental design of a pollination experiment in four trees in a Thuja plicata seed orchard. One hundred percent self-pollen (selfed) and 0% self-pollen (crossed) treatments as well as pollen mixtures with three different ratios of self/cross pollen (25%/75%; 50%/50%; 75%/25%) were applied to each tree. Each "selfed" treatment was performed on three branches per tree and all other pollination treatments were performed on one branch per tree. Pollen FEMALE PARENT (genotype at the G6pd locus) parents 431 (Aa) 432 (AA) 181 (aa) 395 (aa) 431 Selfed Crossed Crossed Crossed 432 Crossed Selfed Crossed Crossed 181 Crossed Crossed Selfed Crossed 395 Crossed Crossed Crossed Selfed 25/75 25/75 431/181 50/50 50/50 75/25 75/25 25/75 25/75 432/181 50/50 50/50 75/25 75/25 25/75 25/75 431/395 50/50 50/50 75/25 75/25 25/75 25/75 432/395 50/50 50/50 75/25 75/25 50 On each tree, 12 different treatments' were randomly assigned to bagged branches while the two remaining branches served as controls. Each tree received three self-pollen treatments and three outcross-pollen treatments (one from each of the three other parent trees). Pollen mixtures with three different ratios of self to outcross pollen were also applied (Table 5.1). Each treatment was performed two to three times in early March after female cones on a branch had become receptive as indicated by the presence of pollination drops on the micropyle (Colangeli and Owens, 1990). To pollinate a branch, a small hole was made in the isolation bag, the nozzle of the squeeze bottle was inserted and a few puffs of pollen were blown in the bag. The bag was shaken to distribute pollen evenly, and the hole was sealed with tape. In June the paper bags were removed and replaced with mesh bags to protect the cones from insect damage. In late summer, once the cones were dry and had turned brown, the branches were removed from the trees. Seed viabUity - To determine seed set for each treatment, 100 seeds from each branch were cut in half with a razor blade, blind to the identity of the treatment. Only pollinated ovules in redcedar developed into seeds (Owens et al, 1990). Full seeds had a plump white embryo, while aborted seeds had a shriveled empty, brown embryo. I use the term inbreeding depression to describe the decrease in self-fertility even though this could be due to late acting self- incompatibility mechanisms (Seavey and Bawa, 1986). For each tree, /, inbreeding depression (5) at the seed stage was calculated as: b = l-(SJSJ where Sai is the proportion of full seeds for the self pollination treatment, and Sci is the proportion of full seeds for the 100% outcrossed pollen treatment. An analysis of covariance (ANCOVA) was used to test for the effect of maternal tree and pollination treatment on the proportion of full 51 seeds. This statistical analysis as well as the others reported below were performed using JMP (version 3.2.1, SAS Institute, 1997). The proportion of full seeds partly depends on the proportion of self-pollen received and inbreeding depression. Polyembryony could increase the number of viable seeds if viable embryos rescue ovules that also contain inviable embryos. The expected proportion of full seeds for each tree, i, and treatment, k, in the absence of polyembryony was calculated as: fit = ( l-Pk)Sci + Pk(Sai) where p is the proportion of self-pollen in the treament k, and Sc and Sa are values estimated for tree i. Wilcoxon sign-rank tests were used to test whether there was an increase in the proportion of observed full seeds compared to the expected (/). Embryo competition - In mixed-pollen treatments, enzyme electrophoresis was used to genotype the offspring at the G6pd locus. For each treatment, 100 seeds, when available, were germinated and genotyped (blind to the identity of the parent and treatment) using the methods described in O'Connell et al. (2001) and Chapter 2. When both parents in the pollen mixture were homozygous for alternative alleles, the offspring were homozygous for the parental allele when selfed, or heterozygous when outcrossed. In the heterozygous tree 431, the proportion of selfed seedlings, s, was estimated as s = 2NAA/(NAA + Naa) where NAA is the number of seedlings with the "AA" genotype and Naa, the number of seedlings with the "aa" genotype. The paternal trees in this case always had the "aa" genotype (Table 5.1). When tree 431 was used as the cross-pollen parent in mixed pollinations, the proportion of selfed seedlings was estimated as 5 = l-(NAa/ where NAa is the number of heterozygous offspring and q is the proportion of the alternative allele in the outcross pollen pool (i.e. q is fixed at 0.5). In this case, the maternal trees had an "aa" genotype. Expected seed set and selfing with polyembryony - The proportion of full and selfed seeds expected with one, two or three embryos per ovule were also calculated following Sorensen (1982). This model incorporates estimates of both selfed and outcrossed embryo viabilities as well as the probability of ovules containing different combinations of selfed and outcrossed embryos. Estimates of selfed-embryo viabilities were calculated as ai~\-(\-SaiY where Sai is the proportion of viable seeds set in tree i after controlled self-pollinations and n is the number of embryos in an ovule. Outcrossed embryo viabilities were calculated as q=l-(l-5d)" where Sci is the proportion of viable seeds after controlled cross-pollinations (Sorensen, 1982). This model assumes that if an ovule contains at least one viable embryo, it will set a seed. An ovule contains n embryos from four different classes with probabilities Pu P2, P3 and P4. The four classes with their probabilities are: selfed and viable (P, = apk), selfed and non-viable (P2 = (1 - a) pk), outcrossed and viable (P3 = c (l-pk)), and outcrossed and non-viable (P4 = (1 - c) (1 - pk)). The array of embryo classes are multinomially distributed such that P(N{ = nv...,N, = nA) = x p,"' ...p* n,!...n4! where /V„ N2, N3 and N4 are the number of each respective embryo class and N, + N2 + N3 + N4 = n, the number of embryos per ovule. The proportion of full seeds expected in a tree (ft), for each value of n is equal to the probability of an ovule containing at least one viable embryo (Table 5.2). The proportion of selfed seeds expected (s,) will be equal the proportion of ovules with selfed seeds divided by the 53 proportion of full seeds. When outcrossed embryos always outcompete selfed embryos within an ovule, only ovules that contain at least one viable selfed embryo and no viable outcrossed embryo will give rise to selfed seeds (Table 5.2). If selfed and outcrossed embryos have an equal chance of occupying an ovule (chance), then the increase in the expected selfing rate depends on the proportion of ovules with both viable selfed and viable outcrossed embryos. When the winning embryo is determined by chance and when n = 2 the expected increase in the proportion of embryos giving rise to selfed seeds is l/2(2PlP3), and the increase in the proportion 2 2 of embryos with selfed seeds when n = 3 is 2/3(3P, P3) + 1/2(6P,P2P3) + 1/3(3P,P3 ) + 1/2(6P,P3P4) (Table 5.2). Table 5.2 Probabilities of setting a full seed (f,) and setting a selfed seed (s,) with one, two or three embryos per ovule (n). Outcross wins, the ouctrossed embryos always outcompetes the selfed embryos within an ovule. Chance, both selfed and outcrossed embryos have equal chance of occupying an ovule. Abbreviations of probabilities are in the text. Proportion of full seeds expected: n = 1 /= P, + P3 2 2 n = 2 /=1-(P2 + 2P2P4 + P4 ) 3 3 2 2 n = 3 /= 1- (P2 + P4 + 3P2 P4 + 3P2P4 ) Proportion of selfed seeds expected: n=l 5, = P1/(P1 + P3) Outcross wins 2 2 2 n = 2 s, = [P, + 2P,P2 +2P,P4] / [1- (P2 + 2P2P4 + P4 )] 3 2 2 2 2 3 3 2 2 n = 3 s,= [P, + 3P, P2 + 3P, P4 + 3P,P2 + 3P,P4 + 6P,P2P4] / [1- (P2 + P4 + 3P2 P4 + 3P2P4 )] Chance: 2 2 2 n = 2 s,. = (P, + 2P,P2 +2P,P4 + P,P3) / [1- (P2 + 2P2P4 + P4 )] 3p 2 p + 3p 2p 3P 2p 3p p 2 6P P 2p 2p 2 n = 3 x i 2 + i 4 + A + i 4 + A 4 + i 3 + 3/jP2P3 + PrP3 + 3PtP3P4 3 3 2 2 " 1 - (P2 + P4 + 3P2 P4 + 3P2P4 ) 54 Wilcoxon sign-rank tests for each treatment were used to test whether the observed proportion of selfed seedlings differed from the proportion expected when n = 1. Figure 5.1 illustrates the expected number of self-fertilized seedlings for n = 1, 2 or 3, with no inbreeding depression causing the death of an embryo, when (1) an outcrossed embryo always outcompetes selfed embryos or (2) selfed or outcrossed embryos have equal chances of filling a seed. An ANCOVA was used to test for a difference in slopes for the proportion of expected vs observed selfed seeds with different proportions of self-pollen. Fitness of self-pollen - Self-pollen can perform differently depending on its frequency in the pollen pool. The fitness of self-pollen relative to outcross pollen was calculated as: w =PkWs + (l-pk)w0 where, w is the total fitness, pk is the proportion of self-pollen applied and 1 - pk, the proportion of outcross-pollen applied. The proportion of selfed seeds observed, s, is therefore: s=PkWs/\pkws + (l-pk)w0] To obtain the fitness of self pollen, vv,, relative to outcross pollen, let w0 = 1 s=PkwJ\pkws + {\ -pk)] which when solved for the relative fitness of self pollen, ws, gives ws = s(l -pk)lpk(\ -s) 55 1.00 n = 1 n = 2, chance co 0.80 n = 2, outcross wins T3 n = 3, chance o 0.40 o Q. 2 °- 0.20 0.00 0.00 0.20 0.40 0.60 0.80 1.00 Proportion of self pollen Fig. 5.1 The proportion of selfed seeds expected with different proportion of self-pollen and numbers of embryos (n) within an ovule. Two different outcomes are shown: the outcrossed embryo always outcompetes the selfed embryo (outcross wins) or self and outcross embryos are equally competitive (chance). All embryos are viable within an ovule (no inbreeding depression). 56 Results Seed set - In bags receiving no pollen, cones began to develop but remained small and no seeds were produced, indicating a low chance of pollen contamination. Some of the branches were broken by wind over the summer, and therefore some replicates are missing. A total of 4954 seeds were checked for an embryo. Three of the trees, 181, 395 and 432, set a large number of full seeds under all pollination treatments and inbreeding depression was around 30% (Fig. 5.2; Table 5.3). However, in tree 431 the proportion of full seeds decreased strongly with an increasing amount of self-pollen, and inbreeding depression at the seed stage was 93%. An analysis of covariance showed that the proportion of full seeds varied among trees (F = 7.88, df= 3,P = 0.0009) and among treatments (F = 5.68, df= 4,P = 0.0025), but the interaction between tree and treatment was not significant (F = 1.75, df = 12, P = 0.12). On average, trees set 78% of their seeds when three of the trees (181, 395, 431) were used as pollen parents in the 100% outcrossed treatments. Branches pollinated with tree 432 as an outcross pollen parent set only 58% of their seeds. The difference in siring success among trees was not significant (F = 1.872, N = 10, df= 3, P = 0.24). However, the power to detect a difference was low (Power = 0.28). There was no association between the proportion of full seeds observed and the outcross pollen parent used in mixed pollinations. Although tree 431 showed very low self-fertility, it did not show a decreased siring success in outcross and mix-pollen treatments when used on unrelated trees. 1.00 —^—181 --«--395 i . -• -431 0.80 x 432 • •; % J?\ V \\ V J . «^ '^i ^0.60 ..3 o 0.40 o \ Q. 2 • \ CL 0.20 0.00 0.00 0.25 0.50 0.75 1.00 Proportion of self pollen Fig. 5.2 The proportion of full seeds obtained in four Thuja plicata trees with different proportions of self-pollen applied. N = 495. 58 Table 5.3 The proportion of full seeds (SE) and inbreeding depression at the seed stage in four western redcedar trees (181, 395, 431 and 432). N, number of seeds sampled. Outcross treatment Self treatment Tree N Full seeds - Sa N Full seeds - Sa Inbreeding Depression 181 300 0.707 (0.026) 200 0.515 (0.035) 0.27* 395 100 0.870 (0.034) 300 0.600 (0.028) 0.31* 431 300 0.753 (0.025) 168 0.051(0.015) 0.93* 432 300 0.710(0.026) 300 0.527 (0.029) 0.26* *Inbreeding depression was significantly different from zero. P < 0.05. The proportion of full seeds expected, with different numbers of embryos within an ovule, are shown in Table 5.4. The increase in the expected number of seeds with an increase in n, is largest when 50% self-pollen is received. This is when the proportion of ovules with both selfed and outcrossed embryos is at its highest. The tree with the highest inbreeding depression, tree 431, shows the greatest increase in expected seed set between ovules with one and multiple embryos. Only one tree, 432, showed an increase in full seeds as expected after mixed pollinations if polyembryony contributes to increasing seed set (Wilcoxon sign rank test, n-6,T = 9.5, P = 0.031; Fig. 5.2). The three other trees showed no increase in seed set. None of the treatments (pooled over all four trees) showed a significant increase in full seeds with mixed pollinations compared to the seed set expected without polyembryony (Table 5.5). 59 Table 5.4 The proportion of full seeds expected (£) based on the number of embryos per ovule (n) and the proportion of self-pollen applied (pk) in four Thuja plicata trees with varying levels of embryo viability. Number of Proportion of self-pollen Tree embryos (n) 0 0.25 0.50 0.75 1 181 1 0.707 0.659 0.611 0.563 0.515 2 0.707 0.664 0.617 0.568 0.515 3 0.707 0.665 0.619 0.569 0.515 395 1 0.870 0.803 0.735 0.668 0.600 2 0.870 0.816 0.753 0.681 0.600 3 0.870 0.820 0.760 0.687 0.600 431 1 0.753 0.578 0.402 0.227 0.051 2 0.753 0.620 0.459 0.269 0.051 3 0.753 0.633 0.478 0.286 0.051 432 1 0.710 0.664 0.619 0.573 0.527 2 0.710 0.668 0.624 0.577 0.527 3 0.710 0.670 0.626 0.578 0.527 60 Table 5.5 The proportion of full seeds expected (ft) without polyembryony (when n = 1) and observed full seeds for three different proportions of self-pollen in four Thuja plicata trees. N, number of treatments. Proportion of Full seeds Wilcoxon sign-rank test self-pollen fa) N Expected Observed (SE) Difference T P (2-tailed) P(l-tailed) 0.25 8 0.676 0.591 (0.067) -0.085 8 0.30 0.16 0.50 8 0.592 0.660 (0.058) 0.068 7 0.25 0.15 0.75 7 0.548 0.598 (0.096) 0.05 4 0.46 0.27 All treatments 23 0.607 0.617 (0.410) 0.01 9 0.79 0.40 Realized selfing rates - A total of 2553 seedlings were genotyped at the G6pd locus. Tree 431 yielded few germinants, resulting in a large error of selfing estimates. Therefore tree 431 was excluded from further analyses. For the remaining three trees, the proportion of selfed seedlings was significantly larger in the 75% self-pollen treatment than in the 25% self-pollen treatment (Fig. 5.3). The 50% self-pollen treatment did not significantly differ from the other two treatments (Tukey test over all treatments: TV = 18; F = 5.86, P = 0.013). The proportion of selfed seeds was lower than expected in the 50% and 75% self-pollen treatments. In the 25% self-pollen treatment there were more selfed seeds than expected, but this difference was not significant. The proportion of selfed seeds was only significantly lower than expected in the 75% self-pollen treatment (Table 5.6). The slopes for the proportion of selfed seeds observed, and expected when n - 1 and with 28% inbreeding depression, were significantly different (ANCOVA: F = 19.49, df = 1, P = 0.0001; Fig. 5.3). With higher levels of inbreeding depression, fewer ovules should contain both selfed and outcrossed embryos, and the expected decrease in selfing is not as pronounced as when all embryos are viable, especially with low proportions of self-pollen. 61 0.80 -' — n •=: 1, no inbreeding depression 0.70 - n ;= 1 Q- - n •= 2, chance ^ to -• - - n = 2, outcross wins "S 0.60 t n = 3, chance CD CO -t- - h = 3, outcross wins • selfed observed Jj 0.50 CD co o 0.40 co o 0.30 Q. 2 Q_ 0.20 0.25 0.50 0.75 0.10 Proportion of self pollen Fig. 5.3 The proportion of selfed seeds expected and observed (± SE) with different proportions of self-pollen. N = 2028. n, number of embryos per ovule. Embryo viabilities for the expected selfed seeds are based on the mean of three trees. See Fig. 5.1 for more details. 62 Table 5.6 The proportion of selfed seeds expected (when n = 1) and observed for three different proportions of self-pollen in three Thuja plicata trees (181, 395 and 432). N, number of replicates. Proportion of Selfed seeds Wilcoxon Sign-rank self-pollen (pk) N Expected Observed (SE) Difference T P (2-tailed) P(l-tailed) 0.25 6 0.193 0.312(0.061) -0.119 7.5 0.156 0.078 0.50 6 0.419 0.390 (0.041) 0.029 2.5 0.68 0.34 0.75 6 0.683 0.527 (0.026) 0.156 10.5 0.031 0.016 All treatments 0.432 0.410 (0.032) 0.022 17.5 0.468 Success of self-pollen - The relative success of self-pollen changed depending on the proportion applied on a tree (Table 5.7). Self-pollen had the greatest success when it was relatively rare, performing 1.5 times better than outcross-pollen when it constituted 25% of the pollen pool. However when self-pollen was common (75% self-pollen) it showed reduced success with only 0.382 of the relative success of outcross pollen. Self-pollen performed similarly in the three trees, with the highest success in the 25% self-pollen treatment. Because of unequal variance among treatments a Welch ANOVA was used to test for a difference among treatments (SAS Institute, 1997). Mean success of self-pollen was significantly different among treatments (F = 9.192, df= 3, P = 0.0027). When the 25% pollen treatment was excluded means for the other treatments still differed from each other (ANOVA , F = 6.058, df= 2,P = 0.01). 63 Table 5.7 Fitness of self-pollen relative to outcross-pollen (ws) when applied at different proportions in three Thuja plicata trees. Proportion of self pollen applied (pk) Maternal tree 0.25 0.50 0.75 1.00 181 2.083 0.547 0.441 0.729 395 1.366 0.667 0.365 0.690 432 1.148 0.811 0.340 0.742 mean (SE) 1.533 (0.388) 0.675 (0.106) 0.382 (0.038) 0.720 (0.066) Discussion Polyembryony as a rescue mechanism -1 detected no significant increase in seed set in mixed-pollen treatments compared to the seed set expected based on pure self- and outcross- pollen treatments. Because the majority of western redcedar trees have high self-fertility, if polyembryony contributes to an increase in seed set, this increase is probably too small to detect. In fact, Table 5.4 shows expected increases in seed set of less than 1% between n = 1 and n = 3 for two of the trees, 181 and 432. Overall there were only 1% more full seeds observed than expected when n = 1. We would expect trees with the highest level of inbreeding depression to show the greatest increase in seed set. However, contrary to expectation tree 432, which had the lowest level of inbreeding depression (26%, Table 5.3) showed a significant increase in seed set following mixed pollinations. On the other hand, tree 431, which set only 7/200 seeds after pure self-pollinations and had 93% inbreeding depression showed no difference between expected and observed seed set. If polyembryony does play a role in rescuing ovules these differences should only be detectable in trees with low self-fertility but this was not the case for tree 431 in western redcedar. The limited number of trees makes it difficult to make firm conclusions. On the other 64 hand, if embryo death occurs late in seed development, after one embryo has already outcompeted the others, polyembryony should not affect seed set. Embryo competition - The proportion of selfed seedlings decreased significantly only when the highest proportion of self-pollen was applied (pk = 0.75). When low proportions of self-pollen were applied 0A = 0.25) there was actually an increase in the proportion of selfed seedlings, this difference, however, was not significant and a large error was associated with the selfing estimate. With the observed level of inbreeding depression over three of the trees (5 = 28%) the competition among embryos and the relative decrease in selfing should be strongest when 55-65% self-pollen is received (for n = 2 or 3). Sorensen (1982) found similar results in Douglas-fir (Pseudotsuga menziesii) and noble fir (Abies procerd), and stated that self- pollination rates of around 50% should confer the greatest advantage to polyembryony and that at low proportions of self-pollen, polyembryony should contribute little to increasing outcrossing rates. A decrease in selfing is not only due to a higher competitive ability of outcrossed embryos. Figure 5.3 shows that even when selfed and outcrossed embryos have the same competitive ability (i.e. chance) there is still a decrease in the proportion selfed seeds due to lower viability of selfed embryos. Because a delay of almost four months occurs between pollen germination and fertilization of ovules (Owens and Molder, 1980), redcedar has the potential for strong post- pollination selection for outcrossed offspring via differential pollen tube growth rates of self vs outcross pollen. This siring advantage of unrelated over related pollen in conifers would be similar to the cryptic self-incompatibility system found in angiosperms. However, in angiosperms differential pollen tube growth leads to an advantage of very large magnitude for outcross pollen. Bateman (1956) obtained 90% outcrossed seeds after self and outcross pollen were applied in a 1:1 ratio in Cheiranthus cheiri (Brassicaceae), but there was no difference in seed set when each pollen type was applied separately. In this study I obtained outcrossing rates 65 of 61% after a 1:1 ratio of self to outcross pollen was applied, and almost all of the reduction in selfed seeds can be accounted for by differences in self-fertility. Eckert and Allen (1997) found a similar advantage of outcross over self-pollen in Decodon verticillatus (Lythraceae), but only half of it could be attributed to early inbreeding depression. They found that differences in pollen tube growth rates likely contributed to the additional advantage of outcrossing over selfing. Early inbreeding depression vs self-incompatibility - Most conifers exhibit low seed set following self-pollination (Sorensen, 1969; Franklin, 1972; Namkoong and Bishir, 1987; Savolainen et al., 1992). Seed abortion following self-fertilization is thought to be caused by deleterious recessive alleles (Charlesworth and Charlesworth, 1987). In this study, I use the term "inbreeding depression" to describe early embryo death, even though this could actually be a case of late self-incompatibility (J. Owens, University of Victoria, pers. com.) Post-zygotic self- incompatibility has been found in several angiosperm trees and may be occuring in gymnosperms (Seavey and Bawa, 1986). In Pinus taeda, Williams et al. (2001) found that a lethal factor linked to a microsatellite marker coincided with the separation of the embryo from the megagametophyte, and this lethal factor acted in an overdominant fashion suggesting a self- recognition mechanism. This system parallels late self-incompatibility found in some angiosperm trees and may operate in other conifer species that show extremely low seed set following self-pollinations. If such a factor exists in redcedar, however, it must have low potency given the high selfing success that I observed overall. Purging of inbreeding depression - Although one tree in our study showed very low self- fertility, the other three showed levels typical for western redcedar (J. Russell, cited in Husband and Schemske, 1996). It has been suggested that the high level of inbreeding depression, usually 66 associated with conifers, has been purged in T. plicata (El-Kassaby et al, 1994). In order for plants to purge their inbreeding depression inbred plants must survive to reproduce (Lande et al, 1994). Late acting inbreeding depression such as a reduction in growth rate can lead to the death of inbred trees before they reach the reproductive stage. Sorensen (1999) looked at the survival rate of three early successional species of conifers (Abies procera, Pinus ponderosa, Pseudostuga menziesii, Pinaceae) with 10% inbreeding depression in growth rate. Because these species are not shade tolerant, late inbreeding depression can successfully eliminate inbred individuals which will be outcompeted by faster growing outbred trees. Western redcedar has also shown 10% inbreeding depression for growth rate (J. Russell, pers. com.) but unlike the species listed above it occurs at all stages of forest succession and tolerates shade (Minore, 1990). Slower growth caused by inbreeding depression will not necessarily lead to the death of inbred individuals in western redcedar so that they can survive and reproduce. Survival of inbred individuals in western redcedar is reflected in positive inbreeding coefficients at the adult stage in natural populations (Chapter 6). The high level of self-fertility in western redcedar is facilitated by the purging of early inbreeding depression in natural populations. The importance of pre-pollination mechanisms - In this study a large number of self- pollinated ovules survived to the seedling stage in mixed pollinations, especially in the 75% self- pollination treatment. This indicates that the amount of self-pollen received (primary selfing rate) in natural populations of western redcedar is an important determinant of the proportion of selfed seedlings. In a previous study a mean selfing rate of 28.5% over six populations was measured in seedlings (O'Connell et al, 2001; Chapter 2). Based on the results of this study we would expect self-pollination rates in natural populations to be similar to the measured selfing rate. In contrast, the self-fertile conifer, Picea omorika, showed 69% seed set after controlled self- pollinations but very high outcrossing rates (nearly one) in natural populations (Kuittinen 67 and Savolainen, 1992). The authors suggest that pre-pollination mechanisms such as protogyny are responsible for the low levels of natural selfing. In western redcedar, polyembryony can contribute to a decrease in selfed seedlings when levels of self-pollen are high, but in general low post-pollination competition means that pre-pollination mechanisms will play an important role in determining selfing rates in most trees. Other trees that show very low levels of self-fertility, such as tree 431 in this study, should have high outcrossing rates but low seed-set with high levels of self-pollination. A combination of the degree of self-pollination and individual tree self-fertility will determine outcrossing rates in redcedar. Correspondingly, outcrossing rates in redcedar should vary among trees and populations as has been found in other studies (El- Kassaby et al, 1994; O'Connell et al, 2001, Chapters 2 and 4). Based on data from Owens and Molder (1980) which is mostly descriptive, I assumed that two to three embryos per ovule is typical in western redcedar. However, redcedar can potentially produce more embryos per ovule as the number of archegonia per ovule is higher. If outcrossed embryos are preferentially selected, then the decrease in selfing rates should be greater with a larger number of embryos. In this case, the high number of selfed seedlings observed in the low pollen treatment (Pk = 0.25) would seem to contradict the competitive advantage of outcrossed embryos. 68 Chapter 6 Range-wide genetic structure and diversity in western redcedar Introduction The amount and partitioning of genetic diversity in a species is related to life-history and ecological traits but is mostly the result of species-specific history (Hamrick et al., 1992). A review of the plant isozyme literature has shown that long-lived woody perennials have significantly more genetic diversity at the species level than annuals, short-lived perennials and long-lived herbaceous perennials (Hamrick et al., 1992). One long-lived conifer, western redcedar (Thuja plicata Donn ex D. Don, Cupressaceae) is among the least genetically diverse trees. Western redcedar has shown low genetic variation within and among populations for several traits. These include relative amounts of leaf oil terpenes (von Rudloff and Lapp, 1979; von Rudloff et al, 1988), isozymes (Copes, 1981; Yeh, 1988; El-Kassaby etal, 1994), restriction fragment length polymorphisms (RFLP; Glaubitz et al, 2000), and phenotypic traits (Rehfeldt, 1994; Bower and Dunsworth, 1988). A reduction in genetic diversity can occur through a reduction in the effective population size, either through a large historical bottleneck or small, reoccurring bottlenecks during the colonization of new areas. A species-wide bottleneck in Thuja plicata could have led to the low amount of genetic variation observed in present day populations (Yeh, 1988; El-Kassaby et al, 1994). The present range of redcedar extends from southeastern Alaska to Northern California along the coast of western North America, and from southeastern British Columbia to northern Idaho in the interior (Fig. 6.1). The coastal and interior parts of the range are essentially geographically isolated from each other (Minore, 1990). During the last 600,000 years of the Quaternary (2.4 million years ago to the present) a series of glacial cycles of approximately 100,000 years each, separated by warmer interglacial periods of about 10,000 years, have had a profound impact on the distribution of North American conifers 69 (Critchfield, 1984; Hewitt, 2000). Paleobotanical records indicate that western redcedar experienced a severe reduction in range size during the last ice age. Along the Pacific coast, the Fraser glaciation reached its maximum southern extent in Northern Washington during the Vashon stage, 15,000 yBP (years before present). Pollen records suggest that western redcedar was found much further south, in California, and when glaciers began retreating redcedar slowly recolonized the coast, reaching the Puget Sound area 10,000 yBP and Northern Vancouver Island 3,000 yBP (Critchfield, 1984; Hebda and Matthewes, 1984). Western redcedar may have persisted in more than one refugium during the last glacial period. Glacial refugia of western mesic forest probably occurred in more than one location along the Pacific coast, including central California in the south, the Queen Charlotte Islands in the north, and in north-central Idaho in the interior (Brunsfeld et al, 2001). Unlike the previous genetic markers used to study western redcedar, microsatellites show extensive polymorphism, and can be used to retrace a more detailed history of the species (O'Connell and Ritland, 2000; Chapter 3). The present patterns of genetic structure in a species are a combination of both historical events and current gene flow. Events such as the number and location of glacial refugia, and the severity of a bottleneck, will affect patterns of allele frequency distribution. Genetic differentiation among regions following deglaciation can arise through two different processes: primary intergradation and secondary contact. During primary intergradation a front of colonizers occupy newly available habitats through long-distance dispersal and form a leading edge of colonizers which is more likely to provide migrants for subsequent colonization (Hewitt, 1993). Secondary contact involves the post-glacial contact of populations isolated in separate glacial refugia. Primary intergradation and secondary contact can be difficult to distinguish from each other but each process should show different patterns of genetic structure in the zone of differentiation. With secondary contact, clines in allele frequency at different loci should coincide, while with primary intergradation they will not 70 necessarily coincide (Barton and Hewitt, 1985). Furthermore, during secondary contact linkage disequilibrium between loci should also persist for a few generations in the contact zone (Durrett et al, 2000). A severe reduction in the effective population size can leave its mark on patterns of allele diversity for several generations. During a bottleneck, populations lose rare alleles more quickly than common alleles, so that the number of alleles per locus is reduced more quickly than expected heterozygosity. This results in an excess of heterozygosity relative to the number of alleles observed in recently bottlenecked populations (Nei et al, 1975; Maruyama and Fuerst, 1985; Comuet and Luikart, 1996). In contrast, in a population quickly expanding from a small effective population, such as during range expansion, mutations will increase the number of rare alleles, and a heterozygosity deficiency should be observed (Maruyama and Fuerst, 1984; Comuet and Luikart, 1996). Both of these events could have occurred, but a heterozygosity excess should be detectable for a shorter time period following a bottleneck than a heterozygosity deficiency. The window of time during which an excess or deficiency in heterozygosity can be detected will also depend on the mutation rate of the genetic markers used. In this study I sampled trees from throughout the range of western redcedar and used hyper-variable microsatellites to retrace its glacial and post-glacial history to answer two questions: Was there a single refugium or multiple refugia during the last glaciation? Is there evidence of a historical bottleneck in western redcedar? Materials and methods Sample collection - Twenty to 37 Thuja plicata trees per population were sampled from 23 populations (mean 27 trees) for a total of 620 trees. Locations of the sampled populations are in Fig. 6.1 and Table 6.1. For most populations, fresh foliage was collected, transported in liquid nitrogen and stored at -80°C until the DNA was extracted using a modified CTAB method (Appendix IV; Doyle and Doyle, 1987). For four of the populations (BC12, BC13, BC14 and 71 BC15), DNA was extracted from 20 to 60 pooled megagametophytes/tree, which had been removed from germinated seedlings (see Chapter 4). DNA extracted for a previous study was used for populations IDA, OR1, OR2, BC3 and some individuals of BC2 (Glaubitz et al, 2000). Samples from throughout the range of Thuja plicata were grouped into 23 populations. Throughout most of its range western redcedar has a continuous distribution and separation into populations is arbitrary. Some of the populations in this study are composed of trees from different, nearby locations (within 150 km of each other). If there is local genetic structure we expect an increase in the fixation index (Fis) when two sub-populations are grouped. I compared Fis values of grouped and ungrouped trees and found no increase in Fis for any of the grouped populations, indicating no significant substructure among the grouped trees (Balloux and Lugon- Moulin, 2002). 72 Fig. 6.1 Range map and location of 23 sampled populations of Thuja plicata. The shaded areas indicate the range of western redcedar. Filled circles represent populations from the southern clade and open circles, populations from the northern clade. Abbreviations are found in Table 6.1. Table 6.1 Location of 23 sampled populations of Thuja plicata. n, number of trees sampled. Population Location n Latitude (°N) Longitude (°W) BC1 Maple Ridge, BC 30 49°13' 122°35' BC2 Castlegar, BC 37 49° 19' 117°40' IDA Moscow, ID 31 46°43' 116°56' BC3 Revelstoke, BC 34 50°58' 118°12' BC4 Valemount, BC 30 52°49' 119°15' BC5 McBride, BC 30 53°17' : 120° 10' CA Eureka, CA 30 40°48' 124°09' OR1 North Bend, OR 25 43°24' 124°13' OR2 Lincoln City, OR 26 44°57' 124°01' WA1 St. Helens, WA 26 46°20' 122°31' WA2 Port Angeles, WA 27 48°07' 123°25' BC12 Coombs, BC 20 49° 19' 124° 19' BC13 Pemberton, BC 20 50° 19' 122°47' BC14 Yellow Point, BC 20 48°58' 123°49' BC15 Paldi, BC 20 48°47' 123°50' BC6 Tofino, BC 34 49°08" 125°54' BC7 Port Hardy, BC 30 50°43' 127°28' BC8 Bella Coola, BC 30 52°22' 126°46' BC9 Queen Charlotte, BC 20 53°16' 132°04' BC10 Prince Rupert, BC 30 54°19' 130°19' BC11 New Hazelton, BC 30 55°15' 127°40' AL1 Petersburg, AL 20 56°48' 132°57' AL2 Ketchikan, AL 20 55°20' 131°38' Total 620 74 Microsatellite screening -1 screened 620 samples at eight polymorphic microsatellite loci. I carried out PCR reactions and fragment detection on a LI-COR 4200 sequencer (Lincoln, Nebraska) as described in O'Connell and Ritland (2000) and Chapter 3. To ensure that bands were uniformly scored, on every gel I used an allelic ladder of two to three individuals with alleles of known sizes and spanning the range of allele sizes at a locus. Five loci, TP1, TP3, TP4, TP7 and TP9, were composed of simple dinucleotide repeats while TP6, TP8, and TP11 had interrupted, compound or more complex repeat motifs (Table 6.2). The number of dinucleotide repeats were obtained by subtracting the number of base pairs in the flanking regions of the microsatellite plus the length of the tail from the allele length, and dividing it by two (O'Connell and Ritland, 2000). Because locus TP6 contains a hexanucleotide repeat, as well as dinucleotide repeats, the number of repeats used in our analyses does not accurately represent the number of repeats for this locus (i.e. one hexanucleotide repeat equals 3 dinucleotide repeats). Diversity analyses - Heterozygosities, allele frequencies, pairwise multilocus Fst (0, Weir and Cockerham, 1984), inbreeding coefficients (F) and F-statistics were obtained using Genepop version 3.3 (an updated version of 1.2, Raymond and Rousset, 1995). Genepop was also used to estimate P-values from exact tests of departure from Hardy-Weinberg equilibrium using the Markov chain method with 1000 iterations (Guo and Thompson, 1992). 75 Table 6.2 Mean diversity at eight microsatellite loci in 23 populations of Thuja plicata Locus n Repeat motif Repeat N Length in bp A A/P F TP1 620 (CA), 13-34 166-208 17 8.57 0.702 0.124* TP3 614 (TG)„ 8-39 176-238 24 9.70 0.796 0.043* TP4 619 (TG)„ 9-24 280-310 14 6.78 0.604 0.103* TP6 606 (GCUGTUAT 78-111 231-297 30 15.70 0.900 0.063* ATGT)„...(GT)„ TP7 601 (CA)„ 6-32 231-283 22 7.83 0.764 0.169* TP8 578 (CA)JVCG(CA)JV 17-49 202-266 30 11.52 0.842 0.271* TP9 617 (AC)„ 20-65 263-351 42 15.00 0.857 0.040* TP11 613 (CTUCA)„ 15-33 200-236 10 5.48 0.663 0.0491 n, number of trees sampled; Repeat N, number of dinucleotide repeats; bp, range of allele lengths in base pairs; A, total number of alleles; A/P, mean number of alleles per population; Hep, mean expected heterozygosity; F, mean inbreeding coefficient. Exact test of departure from Hardy- Weinberg equilibrium: * P-value < 0.00001, | P-value = 0.055. Phylogeography -1 used Gendist in Phylip 3.573c (Felsenstein, 1995) to calculate Nei's genetic distance (Nei, 1972) among all pairs of populations. The Neighbor program was used to construct a Neighbor-joining tree based on Nei's distances. The tree was visualized with Treeview 1.5 (Page, 1998). The Seqboot, Gendist and Consense programs (in Phylip) were used to generate 1000 datasets and trees to obtain bootstrap values for each branch. To test for significant differentiation among groups of populations, I performed an Analysis of Molecular Variance (AMOVA) with Arlequin version 2.000 (Schneider et al, 2000). To test for isolation by distance, I used the R-Package version 4 (Casgrain and Legendre, 2001) to convert latitudes and longitudes into km between populations, and to perform a Mantel test on the correlation between pairwise Fs, and pairwise geographic distances. Correlations between latitude and mean 76 number of alleles per locus (A), expected heterozygosity (Hep) and inbreeding coefficients (F) were calculated using JMP version 3.2 (SAS Institute Inc, 1997). Clines in allele frequencies - Secondary contact and primary intergradation should show different patterns in clinal allele frequency. If separate clades are the result of different glacial refugia a steep cline in allele frequencies should occur in the zone of secondary contact and clines at different loci should coincide. To identify contact zones, pairwise correlations between latitude and allele frequencies were performed for each of the 189 alleles. Alleles that increased in frequency with latitude were termed northern alleles, and alleles that decreased in frequency, southern alleles. Following Turgeon and Bernatchez (2001) the sum of frequencies of all th northern alleles 0^,N) and the sum of all southern alleles (ptS) at the i locus were calculated for each population. The relative frequency of northern or southern alleles in a population were obtained by dividing p,N or p,S by the mean of for all populations (pf Ipt). The relative frequency of alleles for each locus was then regressed on latitude. Piecewise regressions were used to identify contact zones, as shown by an increase in the slope of allele frequencies on latitude. Linkage disequilibrium -1 used GENETIX version 4.02 (Belkhir et al, 2000) to calculate linkage disequilibrium between pairs of northern or southern alleles at different loci. GENETIX adapts the algorithms from LINKDIS of Black and Krafsur (1985). The program allows the calculations of Dy, the linkage disequilibrium between pairs of alleles i and j (i = allele at locus 1 and j = allele at locus 2). For each population, I calculated D, the mean value of Dtj for all pairs of alleles at the different loci. Bottleneck test - The program Bottleneck (version 1.2.02; Cornuet and Luikart, 1996) was used to test for a historical reduction in the effective population size of western redcedar. 77 This program calculates the heterozygosity expected at mutation-drift equilibrium (Heq) for the observed number of alleles (ka) in a population and tests whether there is a excess or deficiency in expected heterozygosity (He). Different levels of heterozygosity are expected at mutation- drift equilibrium depending on whether the loci evolve under the Infinite Allele Model (IAM), the Stepwise Mutation Model (SMM) or the Two Phase Model (TPM). Under the IAM each new mutation gives rise to a new allele different from all existing ones (Kimura and Crow, 1964). Under the SMM new mutations are one size larger or smaller than the original allele (Ohta and Kimura, 1973). Microsatellites are expected to evolve under the TPM, a combination of the IAM and the SMM, where most mutations are stepwise and a small number of mutations are multistep (Di Rienzo et al, 1994). Based on the differentiation patterns observed in the phylogeographic analysis, populations were analyzed as three separate groups: the southern clade (excluding California), the northern clade and California. For comparison, the data were tested for an excess or deficiency in expected heterozygosity under all three mutation-models. For the TPM analysis, variance of the length distribution of multistep mutations was set at 30 (Di Rienzo et al., 1994), and the percentage of single-step-mutations at 70% (the program default values). All results are based on 1000 iterations. To test for a deficiency or excess in He, two different statistical tests were performed. The "sign test" has low statistical power but the Wilcoxon sign- rank test can be used with as few as four polymorphic loci (Cornuet and Luikart, 1996). Lower heterozygosity (Heq) is expected under the IAM than under the SMM for the same number of observed alleles so that microsatellites will often show a heterozygosity excess when analyzed under the IAM model. Isozyme data from a previous study of eight populations of western redcedar from southern British Columbia were also analyzed for comparison (Yeh, 1988). Isozymes are expected to follow the IAM so I analyzed the data under this model only. If we assume that the populations are at mutation-drift equilibrium, the Bottleneck test can also identify which mutation model a locus follows (Cornuet and Luikart, 1996). 78 Results Genetic Diversity - A total of 189 alleles were scored at eight loci ranging from 10 to 42 alleles per locus (Table 6.2). Mean expected heterozygosity at each locus ranged from 0.604 to 0.900. Inbreeding coefficients (Fis) were positive and significantly different from zero for all but one locus (Table 6.2). With microsatellites, positive inbreeding coefficients can be the result of null alleles. This is probably the case for locus TP8, which had the largest Fis value and 6.8% of the individuals did not amplify a single band. If we assume that all individuals that did not amplify are homozygous recessive (r) for a null allele, the estimated frequency of null alleles at this locus is p = Vr = V0.068 = 0.26. Locus TP8 was excluded from other analyses below. Other loci still showed a positive inbreeding coefficient even though all 620 individuals produced bands. The results for genetic diversity at seven loci at the population level are presented in Table 6.3. The mean expected heterozygosity over all populations was 0.755 and the observed heterozygosity was 0.692. A total of 23 private alleles were scored. 79 Table 6.3 Genetic diversity in 23 populations of Thuja plicata at seven microsatellite loci Locus TP8 was excluded. Population A/L PA F South BC1 11.3 3 0.831 0.743 0.105* BC2 11.0 1 0.817 0.756 0.077NS IDA 11.1 1 0.733 0.613 0.164* BC3 10.9 2 0.767 0.761 0.009NS BC4 9.7 0 0.755 0.694 0.088* BC5 10.7 1 0.811 0.714 0.123* CA 9.1 1 0.760 0.703 0.078NS OR1 11.3 1 0.775 0.654 0.150* OR2 12.1 5 0.819 0.602 0.271* WA1 10.6 0 0.797 0.659 0.186* WA2 10.9 1 0.806 0.758 0.058* BC12 10.6 0 0.817 0.805 0.013NS BC13 9.9 0 0.825 0.743 0.109* BC14 8.9 1 0.771 0.729 0.056* BC15 9.6 3 0.741 0.707 0.042* Mean south (SD) 10.5(0.1) Total=20 0.788(0.113) 0.709(0.154) 0.102* (0.146) North BC6 10.4 1 0.733 0.693 0.064NS BC7 10.0 1 0.715 0.690 0.037NS BC8 9.9 1 0.721 0.648 0.107NS BC9 8.0 0 0.660 0.631 0.040NS BC10 9.0 0 0.678 0.711 -0.043NS BC11 8.3 0 0.695 0.669 0.035NS AL1 6.4 0 0.670 0.655 0.025NS AL2 7.3 0 0.669 0.580 0.147* Mean north (SD) 8.7(0.1) Total=3 0.693(0.147) 0.659(0.176) 0.052* (0.137) Overall mean (SD) 9.9(0.1) Total=23 0.755(0.134) 0.692(0.163) 0.085* (0.144) A/L, average number of alleles per locus; PA, number of private alleles; Hep, average expected heterozygosity; H0, average observed heterozygosity; F, average inbreeding coefficient; Exact test of departure from Hardy-Weinberg equilibrium *P < 0.005, NS, not significant after sequential Bonferroni correction (Rice, 1989). 80 Genetic structure - Differentiation among populations was low (mean multilocus Fst = 0.062; Table 6.4). Patterns of genetic structure coincided with geographical locations (Fig. 6.2). Overall, bootstrap values were low except for grouping made up of the eight northern populations (six north-western British Columbia populations: BC6-BC11, and the two Alaska populations: AL1 & AL2) from the rest of the populations. These populations from the northern clade are represented by open circles in Fig. 6.1. I will refer to these populations as the "northern" populations and the 15 others as the "southern" populations from now on. Results from an AMOVA also showed that the northern and southern groups were significantly different from each other (FCT= 0.039; P-value <0.0001; Table 6.5). The interior populations did not form a separate clade from the coastal populations and the branch lengths among central populations were relatively short in the dendrogram (Fig. 6.2). Table 6.4 F-statistics at eight microsatellite loci in Thuja plicata Locus F, F,, F, TP1 0.133 0.066 0.190 TP3 0.046 0.064 0.107 TP4 0.117 0.083 0.191 TP6 0.059 0.032 0.088 TP7 0.171 0.048 0.212 TP8 0.280 0.058 0.322 TP9 0.042 0.062 0.101 TP11 0.042 0.057 0.096 Multilocus 0.112 0.064 0.169 Without TP8 0.085 0.062 0.142 4= u c X) CS B CD o CS 43 3. XJ o ro o ro •s ^ co o CS VO Z' o u cj * s O cS CJ -1 *3u E "cS > a o *> '5. CJ CS 5. "3 co O XI °< c, j* ^aH .§y .o* X) CJ X) O hB i CS 4= acj cS CJ VH X) a CJ o Xt CJ CJ 43 .CJ 60 g '&H 3 O oM to CJ B -B 5 ~ cS s 3 PH o o PH PH OH cn 3 CM CO to M O OH > CJ CS ^ X! B B 60 cj B OH PH < CJ o 43 43 B cS 60 CJ es Q CM g NO CJ 4= NO 0X1 cj 60 82 Table 6.5 Analysis of molecular variance (AMOVA) of the effects of populations and groups (North vs South) on the distribution of genetic diversity in Thuja plicata based on seven microsatellite loci. Source of variation df Sum of Variance Percentage of squares components variation Among groups 1 69.875 0.10917 3.89 Among populations within groups 21 177.437 0.10911 3.89 Within populations 1217 3148.675 2.58724 92.22 Total 1239 3395.986 2.80553 Fixation Indices 2-tailed P FCT (Groups to Total) 0.03891 <0.00001 FSC (Populations to Group) 0.04047 <0.00001 FST (Populations to Total) 0.07781 <0.00001 Isolation by distance - Overall, pairwise genetic distance (0) between populations increased linearly with geographical distance (Mantel's t = 7.206, r = 0.788, P-value <0.00001, N = 253). This relationship was found both among populations within a region (North or South) and between populations from different regions (Pairwise correlations: North vs North: r = 0.412 P < 0.029, N = 28; South vs South: r = 0.737, P < 0.0001, N = 105; North vs South: r = 0.749, P < 0.0001, N = 120; Fig. 6.3). The slope between genetic and geographical distance differed among groups and was significantly steeper in the southern populations than the northern populations (ANCOVA: F-ratio = 5.95, P = 0.016; Table 6.6). 83 2000 2000 1000 1500 2000 Kilometres Fig. 6.3 Pairwise genetic distance (0) as a function of geographic distance between populations of Thuja plicata. Distances are shown between northern populations (N vs N), southern populations (S vs S), and between northern and southern populations (N vs S). 84 Table 6.6 Analysis of covariance (ANCOVA) of the effects of geographical distance between populations and groups (N vs N, S vs S, and N vs S) on pairwise genetic distance (9). Whole model: R2 = 0.701, N = 253, F - ratio = 62.73, P< 0.0001. df F - ratio MS P - value Distance 1 75.4394 <0.0001 Groups 2 12.4201 <0.0001 Distance x Groups 2 6.9467 0.0012 Model 5 115.673 0.028353 <0.0001 Error (residual) 247 0.000245 Northern vs southern populations - Pairwise genetic distances were lower between populations from the northern clade (mean 6 = 0.027 ± 0.013 SD; range 0.0075-0.0529, N = 28; Appendix VI) than between populations from the southern clade (mean 0 = 0.043 ± 0.023 SD; range 0.0043-0.1137). The mean genetic distance between populations from different clades (mean 9 = 0.066 ± 0.016 SD; range 0.0309 - 0.111; N = 81) was significantly larger than between populations within a clade (t = 10.73, df= 251, P < 0.0001). Genetic diversity in northern populations was lower than in southern populations. Both mean expected heterozygosity (r =-0.556, N = 23,P = 0.0059) and the mean number of alleles per locus (r = -0.626, N = 23, P = 0.0014) decreased with latitude (Fig. 6.4). This decrease in diversity was largely driven by differences in genetic diversity between the northern and southern groupings. As a group, southern populations had both significantly more alleles per population (P < 0.0005) and higher mean expected heterozygosity (P < 0.0001) than northern populations (Table 6.3). Seventy-three percent of southern populations (11/15) contained between one and five private alleles while only 37% of the northern populations (3/8) contained one private allele. 85 Mating system - All populations but one (BC10) had a positive inbreeding coefficient and the mean inbreeding coefficient (F = 0.085) was significantly different from zero (Table 6.3). The inbreeding coefficient was negatively correlated with latitude (r = -0.451, N = 23, P = 0.031; Fig. 6.4) and it was not significantly different from zero in seven out of eight northern populations (Table 6.3). Allele size distribution - Only one locus (TP11) out of the eight loci did not show a bimodal or multimodal distribution in allele sizes (Fig. 6.5). There was an overlap in allele size distribution between Northern and Southern populations with the most frequent alleles at a locus often being the same in both groups. Sixty-two alleles were only found in the southern populations while four alleles were found only in northern populations. All these alleles occurred at frequencies below 1% except two alleles at locus TP8 with respective frequencies of 2.7% and 1.7% over all southern populations. 86 13 •12 • 11 - •,* • 9• • • A/bo - • 9 • o; .... o 8 o 7 o o H 'i 0 40 42 44 46 48 50 52 54 56 58 0.82 • • • 0.78 H • - • • • ep • 0.74 • .•: o o o 0.70 • :.o o ... o o 0.66 i i i i 40 42 44 46 48 50 52 54 56 58 40 42 44 46 48 50 52 54 56 58 Latitude (*N) Fig. 6.4 Average number of alleles per locus (A/L), mean expected heterozygosity (Hep) and mean inbreeding coefficients (F) in 23 populations of Thuja plicata as a function of latitude. Filled circles represent populations from the southern clade and open circles, populations from the northern clade. r- uo cd 1.2 £ « CM « -a "S B 33 cd cj I OH H a I-H 43 OH H a 4) PH o o -2 — 11 cd w <« •03) ^2 o CU PH « P > 43 O O Mm 0) cd ^ > ° s§ 0 "3 a u oo 3 43 II 43 O " •C cd c 3 U ^ •» In « ^u I .oa a -3 1 an, • CM .. > O CO T3 D H § ^ a [-1 u to -a *w fig. B 33 ON « 8 S OH -3 •ft 0H0 OH H S3 f—I co .y O 1 s 2 3 •a "a B CN 3 00 42 ll i5 B u« .-a T3 JS _8 pS-i E ° 1 6 £ 43 o 5 4) W 4) CS ^ , , -H «3 43 4) u > *Tt 4) O J£ o H _43 00 ° ^ 2 •H C3 rj- 3 43 II • C ccj C a 43 ^ W t— tfl CO f 1 N — B J2 B U 3 If- Ipupnb^i f jailer -7 6 43* « « t; • 2H 43 89 Clines in allele frequencies - Seven loci showed at least one allele that was significantly correlated with latitude. Twenty-three alleles (i = 7 loci) decreased in frequency with latitude (southern alleles) and 12 alleles (/ = 6 loci) increased in frequency (northern alleles; Table 6.7). When the relative frequency of northern alleles (pt I pt) was plotted on latitude there was a steep cline in allele frequency along the coast that corresponded with the five Vancouver Island populations (Fig. 6.6a). The slope of p, /p, on latitude for the Vancouver Island populations ((3 = 0.42; r2 = 0.325; N = 30; P = 0.001; 95% CI = 0.186, 0.655) was significantly different from the slope for either the southern populations ((3 = 0.044; r2 = 0.179; N = 12;P = 0.0002; 95% CI = 0.021, 0.066; ANCOVA: F = 14.45; N = 102; P = 0.0003), or the northern populations (p = 0.018; r2 = 0.002; N = 36; P = 0.77; 95% CI = -0.106, 0.141; ANCOVA: F = 8.126, N= 66; P = 0.006). This same cline did not occur in the interior populations at the same latitudes. A steep cline in the relative frequency of southern alleles occurred between the California and Oregon populations, much further south than the cline in northern alleles (Fig. 6.6b). The slope of relative allele frequency on latitude for the four southernmost populations (P = -0.352; r2 = 0.388; N = 28; P = 0.0004; 95% CI = -0.530, -0.174) was significantly steeper than for the 19 other populations (p = -0.080; r2 = 0.200; N = 133; P <0.0001; 95% CI = -0.107, -0.052; ANCOVA: JV= 161; F = 25.243; P < 0.0001). The mean relative frequency of southern alleles was significantly larger in the California population than in all other populations (Tukey test; a = 0.05 over all comparisons for all population pairs). There was no significant linkage disequilibrium between the southern alleles, nor the northern alleles in any of the populations, and populations in the contact zone did not show higher mean linkage disequilibrium (D) values than other populations. 90 Table 6.7 Alleles with frequencies significantly correlated with latitude, at the 0.05 level, at seven microsatellite loci in Thuja plicata. Allele = number of repeats, r, Pearson correlation coefficient. Locus Allele Frequency r Allele type TP1 17 0.053 -0.5434 south TP1 18 0.44 0.8021 north TP1 19 0.168 -0.6034 south TP1 20 0.084 -0.4201 south TP1 21 0.044 -0.6863 south TP3 8 0.032 -0.5921 south TP3 16 0.103 -0.5351 south TP3 18 0.274 0.5501 north TP3 24 0.067 0.5692 north TP6 84 0.069 -0.4688 south TP6 90 0.032 -0.5468 south TP6 101 0.0103 0.6697 north TP7 12 0.001 -0.5088 south TP7 16 0.327 0.4538 north TP7 17 0.169 -0.4154 south TP7 18 0.017 -0.5593 south TP7 20 0.215 0.6868 north TP7 21 0.101 -0.7650 south TP8 18 0.011 -0.6099 south TP8 25 0.060 -0.5761 south TP8 31 0.199 0.5083 north TP8 46 0.056 0.5252 north TP9 25 0.008 -0.5619 south TP9 27 0.173 0.7072 north TP9 28 0.068 0.4719 north TP9 31 0.06 -0.4512 south TP9 34 0.036 -0.4285 south TP9 35 0.036 -0.4153 south TP9 36 0.125 -0.5174 south TP9 37 0.143 0.6020 north TP9 43 0.006 -0.4144 south TP9 53 0.004 -0.4467 south TP9 58 0.005 -0.4380 south TP11 26 0.107 -0.4232 south 91 Fig. 6.6 Relative frequency of (a) 12 northern and (b) 23 southern alleles as a function of latitude in 23 populations of Thuja plicata. Filled circles represent mean allele frequencies for southern populations and open circles represent means for northern populations. Error bars represent standard error of the mean. Dotted lines represent steep clines in allele frequencies, and populations with crosses are part of those slopes. 92 Bottleneck test - Table 6.8 summarizes the results of the Bottleneck test. Under the Infinite allele model (IAM) the southern group showed a heterozygosity excess at all eight loci and the California population at seven loci. In the northern group, half of the eight microsatellite loci showed a heterozygosity excess and the other half a deficiency. In contrast, under the Two- phase model (TPM) both the northern and southern groups showed a heterozygosity deficiency at seven of the eight loci, and significantly differed from the TPM mutation-drift equilibrium as shown by both the sign test and the Wilcoxon test. The California population however did not differ from the expected TPM equilibrium and only half of the loci showed a heterozygosity deficiency. Under the Stepwise mutation model (SMM) a higher heterozygosity at mutation- drift equilibrium {Heq) is expected than for the other two models for the same number of alleles (Cornuet et Luikart, 1996). The results for the SMM are not shown but all groups showed a significant heterozygosity deficiency {He < Heq) under this model. For isozymes, four of the five loci showed a heterozygosity deficiency under the IAM. This deficiency was not, however, statistically significant due to the small number of polymorphic loci available. Recently bottlenecked populations should also show a mode-shift distribution in allele frequencies so that alleles in low frequency classes (<0.1) become less abundant than intermediate and high frequency classes (Luikart et al., 1998). The Bottleneck program did not show any significant mode-shift toward higher allele frequency classes in redcedar. In fact most alleles (>70%) in both southern and northern populations were found in the lowest frequency class (<0.05; Fig. 6.7). Only the northern group had alleles that had frequencies of over 60%. These were allele 16 at locus TP1 (Fig. 6.5a) and allele 16 at locus TP4 (Fig. 6.5c). 93 0 7-1 Allele frequency class Fig. 6.7 Proportion of alleles at eight loci in different frequency classes for southern populations (filled bars) and northern populations (open bars) of Thuja plicata. (n = 189 alleles). Tt ON cu »H CU CU ON cu ON B V ON ON Tt TJ O d 1 B 3 O T3 to B X cy CU fi TJ ON CU T3 O o cy ON o tn > o o oo V o o 43 TJ "33 A O d d 3 cu TJ 43 a" cj O O CJ B CU CO cy c CU o CO 3 rt o o Tt CU o 43 s fi TJ c d d d 3 PH bp o o 6 CO PH c-i O CU o c/) o CJ 43 60 CJ N TJ X o 3 i cs CU CU o cu o Tt CU r- 00 CD r— 43 &: Tt 6 •8 43 OO 3 O co 55 i- o PH <2 60 >. 'rt 43 N cy ON o cu 'S ^ ON B o o ON CU O co 00 *-» v o IT) o CU 3 cu 43 PH B d C*H O cy .5 PH TJ X 1 A CU o 3 CO cy B fi B o TJ 1T5J <433 'rrtt CU CO ON CU O TJ A ON CO 43 CJ cj O o d cu B E PH CJ e X CU CU 'u o CN cy WH JD PH CJ NO ON CO X TJ 60 G 00 ON CJ 1- O s o CO o co 43 OH -3 "rt a o O .60 d d T> O 3 B 00 CJ PH O CJ O cj CJ X <2 "rt 43 OH CJ a O CU X fi o cu CJ CJ co rt 43 x o .60 CU X 4^ 'S NO Tt 00 1 G cj cj o 00 00 3 43 B o CU u 00 B E Tt -M CO CU 60 3 CU 3 o .rt CJ cu rt TJ X OH O CU cu CU o 6o -a CJ PH PQ °H ^ 'S cu 4? cj X CO CU 43 o 43 cu 60 O CJ O >, lH rH CO CN 00 N PH &: *3 i—i o IT) CN O rt Tt NO u § >> TO 4—» CU CU T3 O CU B 43 CJ O in rt a rt -4—* 00 o CU rt * 43 NO 43 B E E cy HS T) cu CU _3 Ji 43 43 I '5 hC rt "u 3 » t! N X ca 3 O cu O I O o rt W H 60 U «: PH CM oo U # 95 Discussion Phylogeographic structure - The distribution of genetic diversity in western redcedar suggests the presence of at least three refugia during the last glacial period. Three main patterns stand out from the range-wide phylogeographic analysis. First, the northern populations along the coast of British Columbia and Alaska form a distinct clade from the southern coastal and interior populations. Second, the California population is divergent from all the other populations. Third, the interior and coastal populations, although geographically disjunct, are not genetically distinct. The steep cline in the frequency of northern alleles and the greater genetic distance of populations from different clades, compared to populations within a clade, support Vancouver Island as a zone of secondary contact between trees from southern and northern refugia. The population BC6 near Torino appears to be composed of a mixture of alleles from the northern and southern clades. A probable refugium for the northern populations of western redcedar is near the Queen Charlotte Islands, British Columbia (Fig. 6.8). The strongest evidence for a northern glacial refugium of mesic forests off the coast of the Queen Charlotte Islands are 16,000 year old fossils of several plant species, including Sitka spruce (Picea sitchensis) (Warner and Mathewes, 1982). Sea levels were lower during the last glacial period and forests would have persisted in an area now under water. The occurrence of a north/south split on Vancouver Island coincides with the results of a previous study where restriction fragment length polymorphisms (RFLP) were used to infer the genetic structure of western redcedar over its range (Glaubitz et al, 2000). In this RFLP study, trees from southern coastal British Columbia were pooled with trees from northern Vancouver Island. The results showed that the southern British Columbia coastal group was as closely related to the Queen Charlotte Islands group, as to the groups from the coastal United States and British Columbia interior. These results are not surprising as the British Columbia coastal group would have included individuals from both the northern and southern clades outlined in this study. 96 A north/south split in genetic structure occurs in several other plant species along the west coast of North America. Using cpDNA markers Soltis et al. (1997) found that populations of six plant species (three herbaceous perennials, a tree, a shrub and a fern) were all geographically structured into northern and southern clades that met in south-central Oregon. They attributed this north/south split to the presence of more than one glacial refugium. The location of the steep cline in southern allele frequencies in western redcedar coincides with the location of the north/south partitioning of the species reviewed by Soltis et al. (1997). The lack of differentiation between interior and coastal populations and the divergence of the California population suggest a third possible refugium, in either the interior in northern Idaho, or along the coast of Oregon (Fig. 6.8). Richardson et al. (2002) suggested an east to west colonization of the coast from a refugium in north-central Idaho for Pinus albicaulis. The possible existence of an interior mesic forest refugium is based on the present disjunct range for several different western plant and animal species (Brunsfeld et al, 2001). No mesic forest refugium has been confirmed, however, south of glacial limits as a source for inland populations because no Quaternary pollen sites have been sampled in this area (Mehringer, 1985). Alternatively, the interior could have been colonized by trees from a coastal refugium. Sampled locations in fossil studies south of the glaciers are also sparse along the coast. However, Gottesfeld et al. (1981) identified late Pleistocene (>35,000 radiocarbon yBP) macrofossils of Thuja plicata from the western Cascade mountains, 55 km east of Eugene Oregon. Figure 6.8 illustrates the possible recolonization routes from three separate glacial refugia. 97 Fig 6.8 Hypothesized post-glacial colonization routes for Thuja plicata from three glacial refugia discussed in the text. (1) California, (2) Queen Charlotte Islands, and either (3) western Oregon or (4) northern Idaho. 98 The similarity in the genetic structure of several co-occurring species will help identify the location of glacial refugia and post-glacial colonization routes (Brunsfeld et al, 2001). The genetic structure of other conifers with a geographic distribution similar to Thuja plicata has been studied using isozymes. These include Sitka spruce (Picea sitchensis, Yeh and El-Kassaby, 1980), lodgepole pine (Pinus contorta, Wheeler and Guries, 1982), western white pine (Pinus monticola, Steinhoff et al, 1983), pacific yew (Taxus brevifolia, El-Kassaby and Yanchuk, 1994), mountain hemlock (Tsuga mertensiana, Ally et al, 2000), and yellow cedar (Chameacyparis nootkatensis, Ritland et al, 2001). In many of these studies, the area sampled was limited to British Columbia and no clear patterns of geographic structure arose from the analyses. Exceptions include Pinus monticola, which showed a north/south split near the California-Oregon border corresponding to the pattern observed in redcedar (Steinhoff et al, 1983) . In Pinus contorta sub. latifolia, northern and southern groups had a contact zone in central British Columbia (Wheeler and Guries, 1982). These results are not supported, however, by chloroplast microsatellite marker data (Marshall et al, 2002). Isozymes revealed different clades for Chamaecyparis nootkatensis (Cupressaceae) suggesting multiple glacial refugia along the Pacific coast (Ritland et al, 2001). It would be particularly interesting to obtain a more detailed genetic study on the post-glacial movement of C. nootkatensis because within the Cupressaceae, fossil pollen among species can not be differentiated (Hebda and Mattewes, 1984) . This would help distinguish whether T. plicata or C. nootkatensis was more likely to be in an area at a particular time following deglaciation. Species sharing a similar range distribution will not necessarily show a similar genetic structure. The rate of colonization of new areas will result in differences in genetic structure. For example, a somewhat different structure from western redcedar was observed with allozymes in Alnus rubra (Betulaceae), where Vancouver Island and Queen Charlotte Island populations diverged from mainland populations (Hamann et al, 1998). The authors hypothesized that trees 99 from a northern glacial refugium colonized Vancouver Island when the sea levels were lower and a land bridge connected the Queen Charlotte Islands to Vancouver Island approximately 12,000 yBP. Trees from a southern refugium later colonized the British Columbia mainland. Alnus rubra, unlike Thuja plicata, is an early successional species that moved into recently deglaciated areas more quickly than redcedar. Population differentiation - The amount of differentiation among populations of western redcedar in this study (multilocus Fst = 0.062) was similar to other gymnosperms as measured by isozymes (mean Gst = 0.073 for 102 species; Hamrick and Godt, 1996). A previous isozyme study covering eight populations from southern and eastern British Columbia yielded a lower estimate of differentiation (Fsl = 0.033; Yeh, 1988). When only populations from the same area covered by the isozyme study were included a similar result was obtained (Fs, = 0.027). Divergence is probably underestimated by Fsl for two reasons: (1) Stepwise mutations in microsatellites lead to high homoplasy, decreasing the magnitude of Fsl (Balloux and Lugon- Moulin, 2002). (2) If individuals of a species are isolated into separate refugia during glaciation and undergo separate bottlenecks, most alleles lost in both groups will be rare alleles. Common alleles should be the same in the different groups and will probably survive the bottleneck and increase in frequency independently (Latta and Mitton, 1999). If low Fsl is due to the same alleles surviving in both refugia, values of Fs, should vary among loci. However, individual locus Fst for redcedar had a narrow range, between 3.2% to 8.3%, indicating that Fsl may not have been too biased by this effect. An argument against primary intergradation and in support for multiple glacial refugia as a cause for differentiation among regions in western redcedar, is that trees are less likely to undergo founder effects than annual plants. Austerlitz et al. (2000) outlined two main reasons why trees show lower differentiation in nuclear markers among populations than annual plants. First, because of overlapping generations in trees, multiple 100 colonizers will reach an area before the first arrivals have begun producing offspring. Second, the long-distance gene-flow through pollen dispersal in wind pollinated trees will decrease differentiation among populations. Reduction in genetic diversity - The results of the bottleneck test indicated that a rapid expansion from a restricted population size occurred in redcedar. Of course, this test can only reveal the dynamics in population growth since the last bottleneck. Nevertheless, it indicates the potential impact of glaciation on a species' level of genetic diversity. Both the northern and southern groupings showed heterozygosity deficiencies at microsatellite loci under the TPM indicating a population expansion. Isozyme loci, which were analyzed under the IAM, also showed a heterozygosity deficiency at four of the five loci, but the overall results were not significant. Isozymes are not as well suited for the bottleneck test because they are less likely than microsatellites to be neutral (Cornuet and Luikart, 1996). This is probably the case for locus G6pd (Glucose-6-phosphate-dehydrogenase) which showed a nearly equal frequency of the two alleles in all sampled populations (Yeh, 1988; El-Kassaby et al, 1994; O'Connell et al, 2001). This is also probably why a heterozygosity excess, instead of a deficiency, was observed at this locus. Unlike isozymes, microsatellites in western redcedar are highly variable. The higher mutation rate for microsatellites, compared to isozymes, can explain why a larger number of alleles per locus was found at this marker (see Chapter 7). Both types of markers showed excesses in rare alleles, although they showed different amounts of genetic diversity. For isozyme, a severe bottleneck probably reduced isozyme diversity to one allele per locus at all but one locus (G6pd), while for microsatellites several of the most common alleles probably survived the bottleneck. This is suggested in the multimodal distribution at seven of the eight microsatellite loci. Initially, a few common alleles would have persisted following a bottleneck and a heterozygosity excess may have occurred until new mutations quickly accumulated, 101 leading to the heterozygosity deficiency presently observed. Although northern and extreme southern populations have not been studied for genetic variation at isozyme loci, they showed reduced genetic variation at microsatellite loci compared to populations at the center of the range. A decrease in diversity with latitude is sometimes an argument in favor of recently colonized area (Hewitt 2000). The decrease in diversity in redcedar could also be the result of a more severe bottleneck in the northern populations or lower genetic diversity in these populations preceding the last glaciation. Timing a species-wide bottleneck - The last glacial period lasted approximately 100,000 years. Assuming a generation time of about 50-100 years, populations of western redcedar would have been isolated from each other for about 1000-2000 generations. The evidence of multiple refugia during the last glaciation, combined with the lack of strong differentiation over the range of western redcedar suggests that if a bottleneck reduced species-wide genetic diversity in western redcedar it predates the last glaciation. Because of the long generation time in western redcedar, only 100-200 generations have occurred since deglaciation and even fewer in recently colonized areas. According to pollen records, western redcedar did not reach Northern Vancouver Island until 3000 yBP (Critchfield, 1984). This means that the northern and southern clades have been in contact for no more than 30 to 60 generations. The last interglacial period (Eemian) occurred between 130,000 - 115,000 yBP and was about the same length as the current warm period (Holocene; Adams et al, 1999; Frogley and Tzedakis, 1999). If a severe species- wide bottleneck occurred during a previous glacial period, preceding the last interglacial, levels of genetic diversity during the Eemian would not have been any higher than the present levels of diversity. It is quite plausible that western redcedar has experienced several dramatic contractions and expansions in range and effective population size, and a severe loss in genetic diversity has occurred more than once. The low differentiation between geographically distant 102 southern and northern populations in microsatellites, RFLP (Glaubitz et al, 2000) and leaf oil terpenes (von Rudloff and Lapp, 1979; von Rudloff et al, 1988) indicates that the separate refugia have probably not persisted for more than one glacial cycle. This is in contrast to a species such as ponderosa pine {Pinus ponderosa) that has differentiated to the level of subspecies, suggesting that different parts of the range have been isolated for more than one glacial cycle (Latta and Mitton, 1999). Inbreeding and genetic diversity - Other conifer species in North America have experienced a reduction in range size during glaciation, but few have shown a reduction in genetic diversity as severe as western redcedar. Exceptions include two conifer species, red pine (Pinus resinosd) and Torrey pine (Pinus torreyana), that have almost no isozyme variation (Fowler and Morris, 1977; Allendorf et al, 1982; Simon et al, 1986; Mosseler et al, 1991; Ledig and Conkle, 1983). Genetic diversity is also associated with a species' mating system. Selfing and mixed-mating species generally show lower genetic diversity than outcrossing species (Hamrick and Godt, 1996; Chapter 1). Unlike most conifers Thuja plicata shows high self-fertility and a mixed mating system (Chapters 2, 4 and 5). I found that populations from throughout the range of western redcedar showed a positive inbreeding coefficient with a mean inbreeding coefficient of F = 0.085. The selfing rate, s, based on such a value would be s = 2(F I (1 + F)) = 0.156. This is actually lower than selfing rates observed in natural populations (s = 29 %, O'Connell et al, 2001; Chapter 2) and seed orchards of western redcedar s = 25% (K. Ritland, unpublished). This indicates that although there is some elimination of inbred individuals after the seed census, many selfed seeds also survive and reproduce as trees. Similarly, red pine (Pinus resinosd), has also shown a severe reduction in genetic diversity at isozyme loci. The species has been sampled through most of its range in eastern North America, and only four polymorphic loci have been observed (Hep = 0.002 based on 27 enzyme systems 103 and 64 loci; Fowler and Morris, 1977; Allendorf et al, 1982; Simon et al, 1986; Mosseler et al, 1991). Like redcedar, red pine also shows a potential link between inbreeding and a reduction in genetic diversity. The high self-fertility in 46 trees following controlled pollinations suggests that selfing is potentially high in natural populations of red pine (Fowler, 1965). A reduction in population size in western redcedar, followed by forced mating with relatives and self- fertilization, and purging of deleterious mutations probably added further to a reduction in genetic diversity at neutral linked loci. 104 Chapter 7 Somatic mutations at microsatellite loci in western redcedar Introduction In plants, somatic mutations, i.e., mutations arising from mitosis, can be a significant source of new genetic variation, both within and between individuals. Genetic variation within individuals offers the opportunity for cell lineage selection (Otto and Hastings, 1998) and could be important in plant defense by creating a mosaic of different environments for insect pests (Whitham and Slobodchikoff, 1981; Antolin and Strobeck, 1985). At the population level, somatic mutations can potentially change allele frequencies (Orive, 2001). Somatic mutations are important in the evolution of plant mating systems, particularly in long-lived species such as forest trees, as they contribute to mutational load and inbreeding depression, favoring the evolution of predominantly outcrossing mating systems (Barrett et al., 1996; Morgan, 2001). Somatic mutations can be detected at either genetic marker loci such as RAPDs (as in aspen clones Populus tremuloides; Tuskan et al, 1996) or at conspicuous morphological loci such as chlorophyll deficiency (as in six species of Cupressaceae, Kom, 2001). Microsatellites, or simple sequence repeats (SSRs), offer a special opportunity to observe and study somatic mutations, as their rate of mutation is several orders of magnitude greater than other DNA markers (Ellegren 2000a). Microsatellites consist of tandemly repeated units of DNA of one to six base pairs, and their tandem nature results in mutations due to replication slippage or slipped- strand mispairing during DNA replication (Levinson and Gutman, 1987). Microsatellites are popular markers in population genetic studies, as they are highly variable and co-dominant. Ultimately, observations of microsatellite mutations will enable more accurate inferences based upon microsatellite mutation models, as such observations provide information about the sizes (change in repeat number) and rates (numbers per mitosis or per generation) of microsatellite mutations. 105 Western redcedar (Thuja plicata Donn ex D. Don: Cupressaceae) is a conifer with a mixed mating system and low isozyme diversity (Yeh, 1988; El-Kassaby et al. 1994; O'Connell et al. 2001). Individual trees can live up to 1,000 years and attain heights of more than 50 m (Minore, 1990). Therefore, western redcedar provides a good opportunity to detect and characterize new microsatellite mutations arising via somatic processes. In this study, I sampled haploid megagametophyte tissue in Thuja plicata to detect mutations in a long-lived plant. An advantage of using megagametophytes to detect mutations is that they are part of the germ line, so that the new mutations are heritable. Observations of microsatellite mutations will provide information on the generational somatic mutation rate and the magnitude of size changes of microsatellite mutations. Materials and methods Estimating mutation rate - Mitotic mutation rates can be estimated by calculating the number of cell divisions leading to a new mutation, but this involves several assumptions such as constancy of cell sizes and the fidelity of apical meristems. In addition, when several samples are made throughout a tree, the uncertain origin of cell lineages leading to different sampled tissues further complicates these calculations. Instead, in this study I employ a simple method to estimate per-generation mutation rates, as opposed to per-mitosis rates. On a per-generation basis, the mutation rate is found by simply observing the frequency of new mutations in seed- producing tissue. This is similar to methods based upon the number of genomes sampled (Schlbtterer et al. 1998; Vazquez et al. 2000; Udupa and Baum, 2001; Vigouroux et al. 2002). The mutation rate per locus per generation is estimated as U = m INLK, where m is the number of mutations observed (number of times that genetic differences among sampled tissues within a tree was observed), N is the number of tissues sampled per tree, L is the number of trees sampled, and K is the number of loci sampled. This estimator is derived as follows. If u is the 106 expected per-generation somatic mutation rate, the distribution of the number of mutations found in a sample are the terms in the expansion of LK(u + (l-u))N. If u is small, only two terms predominate: LK(l-u)N (trees with no mutations) and LKNu(l-u)N'' = LKNu (trees with one sample of N mutant). The estimator is then obtained by equating this latter term to the observed numbers of mutants in the total sample (m), and solving for u. This estimator for per-generation mutation rate makes two major assumptions: (1) that the seed-producing tissue sampled represents the historical average age of reproduction for the species, and (2) that new mutations are identified in an unbiased manner. Regarding (1), trees of average mature age should be sampled. Regarding (2), we need to sample at least two tissues per tree to detect mutational changes. We assume that mutant sectors are sufficiently small such that all samples are not all mutant; collection of tissues at points mutually separated by the largest number of mitoses should minimize this possibility of sampling only mutant tissues. Overall, to the extent that the trees represent the average age of reproduction, this estimator would slightly underestimate the true mutation rate, because of the slight possibility that all samples within a tree were new mutants. It does not matter if a mutant sector was missed, as the expected estimate of the frequency of mutant sectors equals the observed fraction of trees with mutant sectors; it does not matter that all mutant sectors have been sampled, only that they have been sampled in an unbiased manner. We also assume that multiple independent mutations do not occur in the same tree; given the relatively low mutation rate (ca. 1 in 1000) this event should be highly improbable and not significantly affect my estimates. The 95% confidence interval was calculated using the Wilson score method with continuity correction which is appropriate for samples sizes above 400 (Wilson, 1927; Newcombe, 1998). The lower and upper confidence limits were calculated as: 107 2np + z2 -1 - zJz2 - 2 - 1/n + 4p(nq +1) Lower = — 2(n + z2) 2np + z2 +1 + zJz2 + 2 - l/n + 4p(nq -1) Upper = - = 2(n + z2) where n is the sample size, p the proportion of mutations observed (LO, q = 1 - p, and z is the standard Normal deviate associated with a two-tailed probability (Newcombe, 1998). Sample collections - During the autumn of 1999 mature cones were collected from a total of 80 trees in four natural populations in southwestern British Columbia (20 trees/population; populations BC12, BC13, BC14 and BC15, Chapter 6). Western redcedar trees produce seed cones throughout the crown including the lower branches that often reached the ground (pers. obs.) Cones were collected from reproductive trees ranging from 4.8 to 36.8 m in height (average height = 20.1 m + 6.7 SD). In three of the populations, cones were collected from two branches from three different heights (top, middle, and lower) in most trees, but only from the top and lower branches in the shorter trees (Fig. 7.1). In one population cone collections were made from one branch from each of two positions (top and lower). Thus my sampling attempted to minimize the probability that all samples within a tree come from the same mutant sector (if the sector exists). To ensure that trees representative of the average age of reproduction were sampled, I also attempted to sample reproductive trees spanning all heights in a population from the shortest and the tallest. The trees sampled are therefore representative of the range in size of mature redcedar trees in the geographical area studied so that a mutation rate per tree, based on these trees, is reasonable. Seeds were mechanically extracted from cones and stored at 4°C until germination. Seeds were germinated following O'Connell et al. (2001), and haploid megagametophytes were 108 separated from the seedlings. From each collection position ten megagametophytes were bulked. A total of 20 to 60 megagametophytes per tree were sampled, and DNA was extracted using a modified CTAB method (Doyle and Doyle, 1987). This use of megagametophytes effectively captures the somatic tissue just prior to its entrance into the germ line (new mutations arising from meiosis are not detected in these bulks) (See Chapter 4). Microsatellites - Each sample of 10 bulked megagametophytes per branch was genotyped at eight polymorphic microsatellite loci developed for Thuja plicata (O'Connell and Ritland, 2000; Chapter 3). Microsatellite repeat motifs ranged from simple dinucleotide repeats to more complex and interrupted motifs (Table 7.1). PCR reactions and allele scoring on a LI-COR 4200 sequencer (LI-COR Inc., Lincoln, Nebraska) were carried out as described in O'Connell and Ritland (2000) and Chapter 3. Table 7.1 Description of eight microsatellite loci used to genotype 80 Thuja plicata trees from four natural populations. N, number of dinucleotide repeats; bp, range of allele lengths in base pairs; A, total number of alleles detected. Locus Repeat motif N bp A TP1 (CA)N 13-34 166-208 12 TP3 (TG)N 9-36 178-232 15 TP4 (TG)N 10-23 282-308 9 TP6 (GC)N(GT)N(ATATGT)N.. .(GT)N 78-110 231-295 28 TP7 (CA)N 11-29 241-277 15 TP8 (CA)NCG(CA)N 23-49 208-266 21 TP9 (AC)N 20-59 263-341 27 TP11 (CT)N(CA)N 25-33 220-232 7 109 Results Microsatellite mutations - After screening the material at eight microsatellite loci, I found a single new allele at locus TP9. Alleles 281 and 291 were found in the lower and mid part of a tree and alleles 281 and 293 in the top part (Fig. 7.1). The new allele (293) likely arose in the upper part of the tree. To confirm that the new allele was not a PCR artifact all the samples from the tree with the new allele were re-amplified and re-scored from the same DNA extraction. The same allele sizes were observed each time. Locus TPS Seed eotoetion Mid Top position Fig. 7.1 Image of a microsatellite gel showing the genotype at locus TP9 for two different heights within the same tree (left). Two collections of ten bulked megagametophytes were made from three heights in each tree (right). 281 bp = 29 dinucleotide repeats, 291 bp = 34 repeats and 293 = 35 repeats. 110 Type of mutation - The size of the new allele corresponded to an increase in one dinucleotide repeat: from 34 to 35 repeats. The new allele size already existed in the sampled populations and was near the middle of the distribution of allele sizes at locus TP9 (Fig. 7.2). Somatic mutation rate estimate - Two positions were sampled in 42 trees and three positions were sampled in 38 trees. Using the above estimator and its assumptions, the estimated rate of somatic mutation rate is U = m INLK = 1 / ((2 x 42 x 8) + (3 x 38 x 8)) = 1 / 1584 = 6.3 x 10"4 mutations per locus per generation with a 95% confidence interval of 3.0 x 10"5 to 4.0 xlO"3. •.-3JQ;.,r . Mi - I Number I 1 II "i& -I I II Fig 7.2 Allele distribution at locus TP9 over four populations of Thuja plicata (N = 80 trees). The new allele, which increased from 34 to 35 dinucleotide repeats, is indicated by the white box, with the arrow showing the original allele. Ill Discussion Somatic mutation rate -1 observed a single somatic mutation occurring in the upper crown of a redcedar tree. The estimated mutation rate of 6.3 x 10"4 per locus per generation (or 3.1 x 10"4 per allele per generation) is within the expected range of 10"3 to 10"4 mutations per generation generally reported for microsatellites (Ellegren, 2000a). In plants, microsatellite mutations rates have been estimated from mutations accumulated in inbred lines. In maize (Zea mays subsp. mays) the estimated mutation rate for 142 microsatellite loci was 7.7 x 10"4 mutations per allele per generation (Vigouroux et al., 2002). Udupa and Baum (2001) estimated microsatellite mutation rates of 1.0 x 10"2 and 3.9 x 10"3 mutations per allele per generation in two annual varieties of chickpea (Cicer arietinum: Fabaceae). These methods would have captured both somatic and meiotic mutations and thus mutation rates should be higher for somatic mutations only. The actual mutation rate per generation in western redcedar is likely higher then reported here, because the method I used was not sensitive enough to detect meiotic mutations. Although the material sampled was part of the germ line and offered the opportunity to detect meiotic mutations, this was difficult because the megagametophyte material was bulked. Results from Chapter 4 showed that alleles occurring at a low frequency, such as a new mutation arising during meiosis, would likely not be detectable in a bulk of 10 megagametophytes. Strictly speaking, a generation's worth of somatic growth should mean something quite precise if we take the demographic definition of the term "generation" into account. It should reflect the survivorship and fecundity schedule of the population. These values are quite difficult to estimate outside of a controlled setting and for such a long-lived plant. During my sampling I attempted to sample from trees representative of the age structure of reproductive individuals in a population to approximate a generation. 112 Mutation model - Information on the mutation processes of microsatellites will help provide more accurate mutation models for population genetics. Distance measures and timing of evolutionary events depend on accurate mutation models. The observed mutation was stepwise, increasing in size by one base pair (Fig. 7.2). Similarly, of 71 observed microsatellites mutations in maize, Vigouroux et al. (2002) found that changes of a single repeat (83% of mutations) were more common than multiple repeats (17%), and a higher proportion of alleles mutated to a larger allele (79%) than a smaller aller (21%). The same directional biases have been reported in birds and humans (Primmer et al. 1996; Ellegren, 2000b). Microsatellites are known to exhibit extensive homoplasy (unrelated alleles of the same size), and correspondingly the new allele in this study mutated to an already existing allele size in the populations sampled. Different loci, and even different alleles, probably mutate at different rates (Ellegren 2000a). Mutations seem to occur in longer loci or alleles, and at loci with simple repeats vs more complex loci. The locus with the observed mutation, TP9, is one of the most variable loci in western redcedar, with 27 alleles detected in 80 individuals (Table 7.1). In a range-wide study, 41 alleles were detected in 620 individuals at locus TP9 (Chapter 6). Changes in allele size can also be caused by changes in the DNA regions flanking the microsatellite. This is unlikely in this study because the new allele corresponded exactly to an increase in one dinucleotide repeat (two extra base pairs) longer than the original allele. The consequences of somatic mutations in redcedar - Generation time can be anywhere from 30-1000 years in Thuja plicata. A high per-generation mutation rate at microsatellite loci can account for the high diversity at these markers compared to other markers, despite there being relatively few generations since a population bottleneck during the last glaciation (Chapter 6). In a long-lived species such as redcedar, deleterious somatic mutations also have the potential of increasing inbreeding depression. 1 Genetic mosaicism - The new allele was found in megagametophytes collected from branches at the top of the tree, but not in any of the lower collections. Korn (2001) showed evidence of a single apical initial cell in several Cupressaceae species. The mutation probably occurred in the apical cell, and all cells above this apical meristem contained the new allele leading to a large sector with a novel genotype. That trees can be mosaics of cells of different genotypes poses interesting questions about plant defense strategies and plant evolution (Whitham and Slobodchikoff, 1981; Antolin and Strobeck, 1985; Gill 1995). 114 Chapter 8 General discussion and conclusions In previous studies, meta-analyses of the literature have helped outline traits associated with long-lived woody plants, such as low genetic diversity, high inbreeding depression and high outcrossing rates. It is by studying variation of these quantities and their association within a species, however, that will gain a better understanding of the how they are interconnected. To better understand the evolution of selfing in a long-lived plant, Thuja plicata, I have assembled three pieces of a puzzle: its mating system, level of inbreeding depression, and genetic structure. To place western redcedar in a phylogenetic context, I first compared its level of selfing and genetic diversity to other species of conifers (Chapter 1). Although western redcedar seems to stand out among trees in terms of its low genetic diversity and high selfing rates, it is not exceptional among more closely related conifers. In fact, its closest relative, Thuja occidentalis, also showed low genetic diversity and some of the highest levels of inbreeding among conifers (Chapter 1). My study focussed on the evolutionary history of western redcedar, and how it has led to the present mating system, patterns of genetic diversity and inbreeding depression. I have shown that self-fertilization evolved with reduced inbreeding depression and reduced genetic diversity. Main findings Mating system - Previously, outcrossing rates in western redcedar were only available for a single seed orchard population. I obtained estimates of inbreeding for six natural populations of redcedar using isozymes (Chapter 2). I found significant amounts of inbreeding, and population outcrossing estimates ranged from 17% to >100% (weighted mean 71%). Outcrossing rates differed significantly among populations, and the results of this study suggested that ecological differences among populations and among trees within populations 115 were partly responsible for the variation in outcrossing rates. To examine the mating system of redcedar at a finer scale, I developed new highly variable microsatellite markers (Chapter 3). I used microsatellites to study the variation in outcrossing rates within an among redcedar trees (Chapter 4). There was no difference in outcrossing among crown positions within trees. In contrast, individual tree outcrossing rates decreased with tree height in all four populations surveyed. This could be the result of larger trees receiving more self-pollen as their pollen makes up a larger proportion of the surrounding pollen cloud. Unlike many other conifers, redcedar was previously shown to be highly self-fertile (Owens et al, 1990). I set out to test whether post-fertilization mechanisms could increase the proportion and number of outcrossed seedlings from a redcedar tree (Chapter 5). To my knowledge this is the first study that has tested for early embryo competition in a conifer. When a high proportion of self-pollen (75%) was applied, unrelated pollen was favoured, suggesting that embryo competition increased the proportion of outcrossed seeds. There was no evidence that polyembryony increased seed set in redcedar but the power to detect an increase was low. Alternatively, if a late acting self-incompatibility mechanism following embryo competition is responsible for seed death (see Williams et al, 2001) polyembryony should not affect seed set. Inbreeding depression - Inbred individuals can be eliminated at different stages of the life-cycle, either through mechanisms that promote outcrossing or through the reduced viability of inbred individuals due to recessive deleterious mutations. Measures of inbreeding, combined with the proportion of surviving individuals from several life-stages in western redcedar, allowed me to estimate inbreeding depression. At the seed stage, I calculated a measure of self-fertility from the proportion of full seeds following self-pollination vs cross-pollination (Chapter 5). I found that individual redcedar trees varied widely in the proportion of viable seeds set after self- fertilization (Chapter 5). Three out of four trees showed about a 30% reduction in seed set 116 following self-pollination compared to cross-pollination. One tree, however, showed a 93% reduction in full seeds in selfed vs outcrossed treatments, setting only 7/200 full seeds after selfing. At the seedling stage, I obtained an estimate of inbreeding depression based on of the proportion of selfed seedlings, relative to the amount of self-pollen applied (Chapter 5). Estimates of inbreeding depression ranged from -0.533 with 25% self-pollen to 0.618 with 75% self pollen. I also obtained selfing rates (s) measured at the seedling stage in natural populations (Chapter 2) and estimates of the number of inbred individuals that survived to adulthood through population inbreeding coefficients (F) (Chapter 6). By using these last two quantities (s and F), and assuming that they are constant over time, inbreeding depression causing mortality between the seedling and adult stage can be estimated in natural populations. Positive inbreeding coefficients in natural populations indicate that at least some inbred trees survive to maturity. To obtain an inbreeding coefficient of F = 0.085 (Chapter 6) selfing rates of about s = 2(F I (1 +F)) = 16% are expected. However, the measured selfing rate at the seedling stage in natural populations is almost twice that (s - 29%; Chapter 2). The adult inbreeding coefficients obtained are lower than expected if all selfed seeds survive as trees. There is probably elimination in inbred individuals in natural populations following the seed stage during germination, seedling establishment and early growth. Inbreeding depression between the seedling and adult stages is measured as: 5 = 1 - wsl w0 and, ws/ w0 = 2 x [(1 - s) F) I s (1 - F)], where ws is the fitness of selfed trees and w0 is the fitness of outcrossed trees as expressed by the number of surviving individuals at maturity (Ritland, 1990b). The estimate of inbreeding depression (5) causing the death of individuals between the seedling and adult stage is about 55%. Although self-fertility in western redcedar is high for a conifer, a large proportion of the inbred individuals are still eliminated after the seed stage in natural populations. Lethal mutations will cause the death of individuals early in the life cycle while late acting inbreeding depression such as a 10% reduction in growth rate in redcedar (J. Russell, unpublished data) will still eliminate inbred 117 individuals that are eventually outcompeted by faster growing trees. Lastly, I found an heritable mutation that arose in the somatic tissue of a redcedar tree, showing the potential to accumulate a higher genetic load in a long-lived organism through somatic mutations, compared to a short lived organism (Chapter 7). Genetic structure and diversity - By developing highly polymorphic microsatellite markers for Thuja plicata, I developed a tool for addressing patterns of genetic diversity in a species with low genetic diversity at other markers (i.e., leaf oil terpenes, isozymes and RFLPs) (Chapter 3). Current patterns of genetic structure in a species are a combination of past events and current levels of gene flow. By studying the genetic structure of redcedar, I have been able to elucidate historical events which led to a reduction in the level of diversity. Microsatellites confirmed the low inter-population differentiation observed with other genetic markers. A study of range-wide genetic structure with microsatellites also revealed more than one glacial refugium for western redcedar (Chapter 6). This study is the clearest evidence of multiple coastal refugia for a tree species in western North America. Because Thuja plicata is often a dominant species in many coastal forests, other associated species will probably show a similar pattern of genetic structure. The allele frequency distribution patterns over all microsatellite loci showed that western redcedar has been expanding in size from small bottlenecked populations and has accumulated a large number of new alleles through mutation (Chapter 6). The low, range-wide diversity in redcedar, combined with the evidence of multiple refugia, suggest that a species- wide bottleneck probably predates the last glaciation. The evidence of an expanding population suggests the potential for recurrent botlenecks with each glacial cycle. Genetic structure and diversity patterns outlined using microsatellite loci will be affected by the mode and rate of mutation of these markers. I observed a single mutation at a microsatellite locus corresponding to a single-step increase in allele size in western redcedar. I 118 estimated a somatic (mitotic) mutation rate for western redcedar of 6.3 x 10"4 mutations per locus per generation. (Chapter 7). The type and rate of mutation observed followed the pattern expected for microsatellites (Ellegren 2002a). Tying it all together - A reduction in species-wide genetic diversity in western redcedar was probably not due to a single event (i.e., a bottleneck), but the interaction of a reduction in population size with an inbreeding system of mating. Other species of Thuja also share a mixed mating system, but not such a severe reduction in genetic diversity (see Chapter 1). This leads us to believe that inbreeding is not a direct consequence of a bottleneck, but may have further accentuated the decrease in genetic diversity in redcedar. High levels of inbreeding are also found in Thuja occidentalis, the closest related species to T. plicata, as well as other species of the Cupressaceae (listed in Ritland et al, 2001). The high self fertility of Thuja plicata (Chapter 5) and high levels of self fertilization measured in natural populations of T. plicata and T. occidentalis (Chapter 1), leads us to conclude that high levels of inbreeding have contributed to a reduction in genetic diversity in both these species. Ecological traits in western redcedar that differ from other conifers and that could further explain some differences in patterns of inbreeding include shade tolerance and a colonizing life strategy (Minore, 1990). Individual redcedar trees with slower growth due to inbreeding are not eliminated as readily as in other species because they can grow in the shade without dying. This allows some inbred adults to survive to maturity and reproduce. The evolution of a mixed- mating system in Thuja plicata could also be due to the need for reproductive assurance in an uncertain environment. Western redcedar usually occurs in mixed-species stands so trees must be able to self-fertilize when conspecifics are rare. 119 Further research Review of mating systems in trees - The literature review of outcrossing rates and genetic diversity in trees presented in Chapter 1 (and Appendices II & III) could be expanded by including other parameters. Outcrossing rates for several tree species are not available but indirect measures of inbreeding can be obtained through the inbreeding coefficient (F). Ritland et al, (2001) listed a number of species in the Cupressaceae with positive inbreeding coefficients suggesting that mixed mating is common in this family. However, there is no comprehensive review of inbreeding coefficients to test whether positive //-estimates are unusual for trees. The combination of population inbreeding coefficients and outcrossing rates can also be used to estimate inbreeding depression within natural populations if these values are assumed to be constant over time (Ritland, 1990b). However, F also measures population substructure and can be inflated by limited dispersal and family structure. Genetic diversity and a mixed-mating system seem to be traits associated with a longer lifespan and woodiness rather than the taxonomic group in which a species is found (i.e., Gymnosperms vs Angiosperms) (Barrett and Eckret, 1990; Hamrick and Godt, 1996; Barrett et al, 1996). Including more angiosperm trees in a review of mating systems may bring out other patterns associated with inbreeding. Another quantity which is increasingly being reported in studies of mating systems in trees is the correlation of paternity (rp), defined as the proportion of full sibs among outcrossed sibs (Ritland, 1989). Estimates of rp for a few conifer species are listed in Chapter 2, and a more comprehensive review of this quantity in trees would provide information on the source of the paternal contribution to seeds. Glacial refugia - Patterns of microsatellite diversity across the species range provide strong evidence that western redcedar survived in at least three glacial refugia during the last ice age (Chapter 6). To confirm the locations of glacial refugia and zones of secondary contact in 120 western redcedar, uniparentally inherited, non-recombining and highly variable genetic markers such as chloroplast microsatellites (cpSSR) could be used. These markers have been used to retrace post-glacial movement and population dynamics in another genetically depauperate species, red pine (Pinus resinosa; Echt et al, 1998) as well as other conifer species (e.g. Pinus contorta, Marshall et al, 2002). Specific geographic locations to concentrate on in redcedar would be areas of possible secondary contact between different refugia. These include populations along the coast of the British Columbia (between Squamish and Bella Coola on the mainland, as well as on Vancouver Island) and populations in northern California to central Oregon. No cpSSR markers presently exist for Thuja plicata so these markers would first need to be developed. The similarity in the genetic structure of several species with similar geographic distribution patterns will help identify and confirm the location of glacial refugia. The data for western redcedar increases the list of plant species for which range-wide phylogeographic data is available in western North America (Soltis et al, 1997). Other conifer species that share a similar geographic pattern include Sitka spruce (Picea sitchensis), yellow-cedar (Chamaecyparis nootkatensis), western hemlock (Tsuga heterophylla) and pacific yew (Taxus brevifolia) (Burns and Honkala, 1990). Although isozyme data suggest multiple refugia for some of these species, more polymorphic markers can provide a better resolution. Allele distribution - One method that can be used to estimate the time since the last bottleneck is to assume that diversity at all loci was reduced to a single allele (Menotti and O'Brien, 1993). This method may be used with isozymes, but it is unrealistic to assume that only one allele per locus survived a bottleneck at microsatellite loci because of their high variability. It is more likely that a few of the more common alleles survived at most loci. The allele size distribution pattern in redcedar ranged from a simple unimodal distribution at one locus to 121 bimodal and multimodal distributions at other loci (Chapter 6). Multimodal distributions could be a consequence of genetic drift, population bottlenecks or occasional multistep mutations (Two-phase mutation model). Simulations of bottlenecks would be useful for understanding their effect on allele size distribution in western redcedar: for example, to test whether bottlenecks lead to a multimodal distribution in allele frequencies. Mating system at the edge of the distribution - Northern populations showed low inbreeding coefficients compared to more southern populations (Chapter 6). Mating system dynamics at the edge of the range might be quite different from the central populations. Estimates of outcrossing have only been obtained for southwestern British Columbia populations. It would be interesting to see if levels of selfing or inbreeding depression in northern populations differ from populations at the centre of the range. Related species - Many generalizations about conifer mating systems have been predominantly based on species in the genus Pinus (Pinaceae). Other genera and families of conifers may show higher levels of inbreeding, self-fertility and/or lower genetic diversity. More data on other species of Thuja and other members of the Cupressacea may show that redcedar is not an outlier among conifers. Levels of inbreeding have been estimated for very few conifer species outside the Pinaceae, so no conclusion can be reached about the mating system of the Cupressaceae in general. Ritland et al. (2001) listed species in the Cupressaceae that had positive inbreeding coefficients suggesting a mixed-mating system in these species. Although the extremely low levels of self-fertility observed in several conifers are often attributed to genetic load, we might actually be observing late acting self-incompatibility mechanisms (Williams et al., 2001; J. Owens, pers. comm.) The Cupressaceae may lack this hypothesized self-incompatibility mechanism found in the Pinaceae allowing them to inbreed. Testing 122 whether polyembryony promotes an increase in seed set and outcrossing in other groups of conifers may yield very different results from those observed in Thuja plicata. Finally, microsatellites develop for one species can often be used for other related species (Karhu et al, 2000). The microsatellites I developed for Thuja plicata might be used to study the mating system and genetic structure of Thuja occidentalis and other related species. Data on the relationship between mating system and genetic structure in related species will provide information on the longer-term dynamics of these quantities. 123 References Adams, J., M. Maslin, and E. Thomas. 1999. Sudden climate transitions during the Quaternary. Progress in Physical Geography 23: 1-36. Adams, W. T., and D. S. Birkes. 1991. Estimating mating patterns in forest tree populations. In Biochemical markers in the population genetics of forest trees. S. Fineschi, M.E. Malvolti, F. Cannata, and H. H. Hattemer (Eds.). SPB Academic Publishing, The Hague, The Netherlands, pp. 157-172. Allendorf, F. W., K. L. Knudsen, and G. M. Blake. 1982. Frequencies of null alleles at enzyme loci in natural populations of ponderosa and red pine. Genetics 100: 497-504. Ally, D., Y. A. El-Kassaby, and K. Ritland. 2000. Genetic diversity, differentiation and mating system in mountain hemlock (Tsuga mertensiana) across British Columbia. Forest Genetics 7: 97-108. Antolin, M. F., and C. Strobeck. 1985. The population genetics of somatic mutation in plants. American Naturalist 126: 52-62. Austerlitz, F., S. Mariette, N. Machon, P.-H. Gouyon, and B. Godelle. 2000. Effects of colonization processes on genetic diversity: differences between annual plants and tree species. Genetics 154: 1309-1321. Balloux, F., and N. Lugon-Moulin. 2002. The estimation of population differentiation with microsatellite markers. Molecular Ecology 11: 155-165. Barrett, J. W., P. Knowles, and W. M. Cheliak. 1987. The mating system in a black spruce clonal orchard. Canadian Journal of Forest Research 17: 379-382. Barrett, S. C. H., and C. G. Eckert, 1990. Variation and evolution of mating systems in seed plants. In Biological Approaches and Evolutionary Trends in Plants. S. Kawano (Ed.) Academic Press, London, pp. 229-254. Barrett, S. C. H., L. D. Harder, and A. C. Worley. 1996. The comparative biology of pollination and mating in flowering plants. Philosophical Transactions of the Royal Society of London, Series B 351: 1271-1280. Barton, N. H., and G. M. Hewitt. 1985. Analysis of hybrid zones. Annual Review of Ecology and Systematics 16: 113-148. Bateman, A. J. 1956. Cryptic self-incompatibility in the wallflower: Cheiranthus cheiri L. Heredity 10: 257-261. Bawa, K. S. 1974. Breeding systems of tree species of a lowland tropical community. Evolution 85-92. Beaulieu, J., and J. P. Simon. 1994. Genetic structure and variability in Pinus strobus in Quebec. Canadian Journal of Forest Research 24: 1726-1733. Beaulieu, J., and J. P. Simon. 1995. Mating system in natural populations of eastern white pine in Quebec. Canadian Journal of Forest Research 25: 1697-1703. 124 Belkhir, K., P. Borsa, J. Goudet, L. Chikhi, and F. Bonhomme. 2000. GENETIX version 4.02. Laboratoire Genome, Populations, Interactions, CNRS UPR 9060, Universite de Montpellier II, Montpellier, France. Black W. C., and E. S. Krafsur. 1985. A FORTRAN program for the calculation and analysis of two-locus linkage disequilibrium coefficients. Theoretical and Applied Genetics 70: 491- 496. Boscherini, G., G. G. Vendramin, and R. Giannini. 1993. Mating system analysis in two Italian populations of Norway spruce. Journal of Genetics and Breeding 47: 45-48. Boyle, T. J., and E. K. Morgenstern. 1986. Estimates of outrcrossing rates in six populations of black spruce in central New Brunswick. Silvae Genetica 35: 102-106. Boyle, T. J. B., C. Liengsiri, and C. Piewluang. 1990. Genetic structure of black spruce on two contrasting sites. Heredity 65: 393-389. Boyle, T. J. B., C. Liengsiri, and C. Piewluang. 1991. Genetic studies in a tropical pine - Pinus kesiya III. The mating system in four populations from northern Thailand. Journal of Tropical Forest Science 4: 37-44. Bower, R. C., and B. G. Dunsworth, 1988. Provenance test of western red cedar on Vancouver Island. In Western red cedar—does it have a future? N. J. Smith (Ed.) University of British Columbia, Faculty of Forestry, Vancouver, Canada, pp 131-135. Brunsfeld, S. J., J. Sullivan, D. E. Soltis, and P. S. Soltis. 2001. Comparative phylogeography of north-western North America: a synthesis. In Integrating ecological and evolutionary processes in a spatial context. J. Silvertown and J. Antonovics (Eds.) Blackwell Science. Oxford, U.K. pp. 319-339. Burczyk, J. 1991. The mating system in a Scots pine clonal seed orchard in Poland. Annales des Sciences Forestieres 48: 443-451. Burczyk, J., W. T. Adams, and J. Y. Shimizu. 1996. Mating patterns and pollen dispersal in a natural knobcone pine (Pinus attenuata Lemmon.) stand. Heredity 77: 251-260. Burczyk, J., and W. Chalupka. 1997. Flowering and cone production variability and its effect on parental balance in a Scots pine clonal seed orchard. Annales des Science Forestieres 54: 129-144. Burns, R. M., and B. H. Honkala (Eds.). 1990. Silvics of North America, Volume 1. Conifers. U.S.D.A. Forest Service Agriculture Handbook 654. Washington, DC. 675 pps. Butcher, P. A., J. C. Bell, and G. F. Moran. 1992. Patterns of genetic diversity and nature of the breeding system in Melaleuca alternifolia (Myrtaceae). Australian Journal of Botany 40: 365-375. Byers, D. L., and D. M. Waller. 1999. Do plant populations purge their genetic load? Effects of population size and mating history on inbreeding depression. Annual Review in Ecology and Systematics 30: 479-513. 125 Casgrain, P., and P. Legendre. 2001. The R-Package for multivariate and spatial analysis, version 4.0 d6. User's Manual. Departement de sciences biologiques, University de Montreal, Montreal, Quebec. URL: http://www.fas.umontreal.ca/BIOL/legendre/. Chaisurisri, K., J. B. Mitton, and W. A. El-Kassaby. 1994. Variation in the mating system of Sitka spruce (Picea sitchensis): evidence for partial assortative mating. American Journal of Botany 81: 1410-1415. Changtragoon, S., and R. Finkeldey. 1995. Patterns of genetic variation and characterization of the mating system of Pinus merkusii in Thailand. Forest Genetics 2: 87-97. Charlesworth, D., and B. Charlesworth. 1987. Inbreeding depression and its evolutionary consequences. Annual Review in Ecology and Systematics 18: 237-268. Charlesworth, D., and B. Charlesworth. 1995. Quantitative genetics of plants: the effect of the breeding system on genetic variability. Evolution 49: 911-920. Charlesworth, D., M. T. Morgan, and B. Charlesworth. 1990. Inbreeding depression, genetic load, and the evolution of outcrossing rates in a multilocus system with no linkage. Evolution 44: 1469-1489. Cheliak, W. M., J. A. Pitel, and G. Murray. 1985a. Population structure and the mating system of white spruce. Canadian Journal of Forest Research 15: 301-308. Cheliak, W. M., B. P. Dancik, K. Morgan, F. C. H. Yeh, and C. Strobeck. 1985b. Temporal variation of the mating system in a natural population of jack pine. Genetics 109: 569- 584. Colangeli, A. M., and J. N. Owens. 1990. The relationship between time of pollination, pollination efficiency, and cone size in western red cedar (Thuja plicata). Canadian Journal of Botany 68: 439-443. Collevatti, R. G., D. Grattapaglia, and J. D. Hay. 2001. High resolution microsatellite based analysis of the mating system allows the detection of significant biparental inbreeding in Caryocar brasiliense, an endangered tropical tree species. Heredity 86: 60-67. Conkle, M. T. 1987. Electrophoretic analysis of variation in native monterey cypress (Cupressus macrocarpa Hartw.). In Conservation and management of rare and endangered plants. Proceedings of the conference. T. S. Elias (Ed.) Sacramento, CA. pp. 249-256. Copes, D. L. 1981. Isoenzyme uniformity in western red cedar seedlings from Oregon and Washington. Canadian Journal of Forest Research 11: 451-453. Cornuet, J. M., and G. Luikart. 1996. Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144: 2001-2015. Cottrell, J. E., and I. M. S. White. 1995. The use of isozyme genetic markers to estimate the rate of outcrossing in a Sitka spruce (Picea sitchensis (Bong.) Carr.) seed orchard in Scotland. New Forests 10: 111-122. Critchfield, W. B. 1984. Impact of the pleistocene on the genetic structure of North American conifers. In Proceedings of the 8th North American Forest Biology Workshop. R. M. 126 Lanner (Ed.) Department of Forest Resources, Utah State University, Logan, Utah. pp. 70-117. Crook, R. W., and W. E. Friedman. 1992. Effects of pollen tube number and archegonium number of reproduction in Douglas-fir: significance for seed orchard management. Canadian Journal of Forest Research 22: 1483-1488. Davidson, R., and Y. A. El-Kassaby. 1997. Genetic diversity and gene conservation of Pacific silver fir (Abies amabilis) on Vancouver Island, British Columbia. Forest Genetics 4: 85- 98. Denti, D., and D. J. Schoen. 1988. Self-fertilization rates in white spruce: effect of pollen and seed production. Journal of Heredity 79: 284-288. Desponts, M., and J. P. Simon. 1987. Genetic structure and variability in populations of black spruce (Picea mariana) in the subarctic region of northern Quebec. Canadian Journal of Forest Research 17: 1006-1012. DiRienzo, A., A. C. Peterson, J. C. Garza, A. M. Valdes, M. Slatkin, and N. B. Freimer. 1994. Mutational processes of simple-sequence repeat loci in human populations. Proceedings of the National Academy of Sciences, USA 91: 3166-3170. DongMei, Z., L. Yue, S. XiHuan, Z. ChunXiao, L. GuoFeng, and L. Jun. 2000. A preliminary study on the mating system of three different populations within an improvement programme for Pinus tabulaeformis Carr. Journal of Beijing Forestry University 22: 13- 18. Doyle J. J., and J. L. Doyle. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15. Durett, R., L. Buttel, and R. Harrison. 2000. Spatial models for hybrid zones. Heredity 84: 9-19. Echt, C. S., L. L. DeVerno, M. Anzide, and G. G. Vendramin. 1998. Chloroplast microsatellites reveal population genetic diversity in red pine, Pinus resinosa Ait. Molecular Ecology 7: 307-316. Eckert, C. G., and M. Allen. 1997. Cryptic self-incompatibility in tristylous Decodon verticillatus (Lythraceae). American Journal of Botany 84: 1391-1397. El-Kassaby, Y. A. 1999. Phenotypic plasticity in western redcedar. Forest Genetics 6: 235-240. El-Kassaby, Y. A., and B. Jaquish. 1996. Population density and mating system pattern in western larch. Journal of Heredity 87: 438-443. El-Kassaby, Y. A., M. D. Meagher, and R. Davidson. 1993. Temporal variation in the outcrossing rate in a stand of western white pine. Silvae Genetica 42: 131-135. El-Kassaby, Y. A., M. D. Meagher, J. Parkinson, and F. T. Portlock. 1987. Allozyme inheritance heterozygosity and outcrossing rate among Pinus monticola near Ladysmith, British Columbia. Heredity 58: 173-181. 127 El-Kassaby, Y. A., J. Parkinson, and W. J. B. Devitt. 1986. The effect of crown segment on the mating system in a Douglas-Fir {Pseudotsuga menziesii (Mirb.) Franco.) Seed Orchard. SilvaeGenetica35: 149-155. El-Kassaby, Y. A., and K. Ritland. 1996. Genetic variation in low elevation Douglas-fir of British Columbia and its relevance to gene conservation. Biodiversity and Conservation 5: 779-794. El-Kassaby, Y. A., K. Ritland, A. M. K. Fashler, and W. J. B. Devitt. 1988. The role of reproductive phenology upon the mating system of a Douglas-fir seed orchard. Silvae Genetica 37: 76-82. El-Kassaby, Y. A., D. Rudin, and R. Yazdani. 1989. Levels of outcrossing and contamination in two Pinus sylvestris L. seed orchards in northern Sweden. Scandanavian Journal of Forest Research 4: 41-49. El-Kassaby, Y. A., J. Russell, and K. Ritland, 1994. Mixed mating in an experimental population of western red cedar, Thuja plicata. The Journal of Heredity 85: 227-231. El-Kassaby, Y. A., and A. D. Yanchuk. 1994. Genetic diversity, differentiation and inbreeding in Pacifc yew from British Columbia. Journal of Heredity 85: 112-117. El-Kassaby, Y. A., F. C. Yeh, and O. Sziklai. 1981. Estimation of the outcrossing rate of Douglas-fir {Pseudotsuga menziesii [Mirb.] Franco) using allozyme polymorphisms. Silvae Genetica 30: 182-183. Ellegren, H. 2000a. Microsatellite mutations in the germline; implications for evolutionary inference. Trends in Genetics 16: 551-558. Ellegren, H. 2000b. Heterogenous mutation processes in human microsatellite DNA sequences. Nature Genetics 24: 400-402. Erickson, V. J., and W. T. Adams. 1990. Mating system variation among individual ramets in a Douglas-fir seed orchard. Canadian Journal of Forest Research, 20: 1672-1675. Ettl, G. J., and D. L. Peterson. 2001. Genetic variation of subalpine fir {Abies lasiocarpa (Hook.) Nutt.) in the Olympic Mountains, WA, USA. Silvae Genetica 50: 145-153. Fady, B., and R. D. Westfall. 1997. Mating system parameters in a natural population of Abies borisii regis Mattfeld. Annales des Sciences Forestieres 54: 643-647. Fast, W., B. P. Dancik, and R. C. Bower. 1986. Mating system and pollen contamination in a Douglas-fir clone bank. Canadian Journal of Forest Research 16: 1314-1319. Farris, M. A., and J. B. Mitton, 1984. Population density, outcrossing rate, and heterozygote superiority in ponderosa pine. Evolution 38: 1151-1154. Felsenstein, J. 1995. PHYLIP (Phylogeny Inference Package) Version 3.57c. Department of Genetics, University of Washington, Seattle, Washington Ferreyra, L. L., A. Latino, A. Calderon, and C. N. Gardenal. 1996. Allozyme polymorphism in Austrocedrus chilensis (D. Don) Florin and Boutelje from Patagonia, Argentina. Silvae Genetica 45: 61-64. 128 Fowler, D. P. 1965. Effects of inbreeding in red pine, Pinus resinosa Ait. II. Pollination studies. Silvae Genetica 14: 12-23. Fowler, D. P., and R. W. Morris. 1977. Genetic diversity in red pine: evidence for low genie heterozygosity. Canadian Journal of Forest Research 7: 343-347. Franklin, E. C. 1972. Genetic load in loblolly pine. American Naturatlist 102: 262-265. Friedman, S. T., and W. T. Adams. 1985. Levels of outcrossing in two loblolly pine seed orchard. Silvae Genetica 34: 157-162. Frogley, M. R., and P. C. Tzedakis. 1999. Climate variability in northwest Greece during the last interglacial. Science 285: 1886-1889. Furnier, G. R., and W. T. Adams. 1986. Mating system in natural populations of Jeffrey pine. American Journal of Botany 73: 1002-1008. Ge, S., M. X. Wang, and Y. W. Chen. 1988. An analysis of population genetic structure of Masson pine by isoenzyme technique. Scientia Silvae Sinicae 24: 399-409. Giannini, R., M. Morgante, and G. Vendramin. 1991. Allozyme variation in Italian populations of Picea abies (L.) Karst. Silvae Genetica 40: 3-4. Giannini, R., L. Parducci, P. Rossi, and F. Villani. 1994. Genetic structure and mating system of silver fir in the Campolino reserve (North Apennines, Italy). Journal of Genetics & Breeding 48: 335-337. • - Gibson, J. P., and J. L. Hamrick. 1991a. Genetic diversity and structure in Pinus pungens (Table Mountain pine) populations. Canadian Journal of Forest Research 21: 635-642. Gibson, J. P., and J. L. Hamrick. 1991b. Heterogeneity in pollen allele frequencies among cones, whorls, and trees of Table Mountain pine (Pinus pungens). American Journal of Botany 78: 1244-1251. Gill, D. E., L. Chao, S. L. Perkins, and J. B. Wolf. 1995. Genetic mosaicism in plants and clonal animals. Annual Review of Ecology and Systematics 26: 423-444. Glaubitz, J., Y. A. El-Kassaby, and J.E. Carlson, 2000. Nuclear restriction fragment length polymorphism analysis of genetic diversity in western red cedar. Canadian Journal of Forest Research 30: 379-389. Godt, M. J. W., J. L. Hamrick, M. A. Edwards-Burke, and J. H. Williams. 2001. Comparisons of genetic diversity in white spruce (Picea glauca) and jack pine (Pinus banksiana) seed orchards with natural populations. Canadian Journal of Forest Research 31: 943-949. Gomory, D., and L. Paule. 1992. Inferences on mating system and genetic composition of a seed orchard crop in the European larch (Larix decidua Mill.). Journal of Genetics and Breeding 46: 309-313. Goncharenko, G. G., V. E. Padutov, and A. E. Silin. 1993. Allozyme variation in natural populations of Eurasian pines II. Genetic variation, diversity, differentiation, and gene flow in Pinus sibirica Du Tour in some lowland and mountain populations. Silvae Genetica 42: 246-253. 129 Goncharenko, G. G., I. V. Zadeika, and J. J. Birgelis. 1995. Genetic structure, diversity and differentiation of Norway spruce (Picea abies (L.) Karst.) in natural populations of Latvia. Forest Ecology & Management 72: 31-38. Gottesfeld, A. S., F. J. Swanson, and L. M. J. Gottesfeld. 1981. A pleistocene low-elevation subalpine forest in the western cascades, Oregon. Northwest Science 55: 157-167. Guo, S. W., and E. A. Thompson. 1992. Performing the exact test of Hardy-Weinberg proportions for multiple alleles. Biometrics 48: 361-372 Guries, R. P., and F. T. Ledig. 1982. Genetic diversity and population structure in Pitch Pine (Pinus rigida Mill.). Evolution 36: 387-402. Hall, P., L. C. Orrell, and K. S. Bawa. 1994. Genetic diversity and mating system in a tropical tree, Carapa guianensis (Meliaceae). American journal of Botany 81: 1104-1111. Hall, P., S. Walker, and K. Bawa. 1996. Effect of forest fragmentation on genetic diversity and mating system in a tropical tree, Pithecellobium elegans. Conservation Biology 10: 757- 768. Hamann, A., Y. A. El-Kassaby, M. P. Koshy, and G. Namkoong. 1998. Multivariate analysis of allozymic and quantitative trait variation in Alnus rubra: geographic patterns and evolutionary implications. Canadian Journal of Forest Research 28: 1557-1565. Hamrick, J. L., and M. J. W. Godt. 1996. Effects of life history traits on genetic diversity in plant species. Philosophical Transactions of the Royal Society of London: Series B 351: 1291- 1298. Hamrick, J. L., M. J. W. Godt, and S. L. Sherman-Broyles. 1992. Factors influencing levels of genetic diversity in woody plants species. New Forests 6: 95-124. Harder, L. D., and S. C. H. Barrett. 1996. Pollen dispersal and mating patterns in animal pollinated plants. In Floral biology: studies on floral evolution in animal pollinated plants. D.G. Lloyd and S.C.H. Barrett. (Eds.) Chapman & Hall, New York. pp. 140-190. Hebda, R. J., and R. W. Mathewes. 1984. Holocene history of cedar and native indian cultures of the North American Pacific Coast. Science 225: 711-713 Hewitt, G. M. 1993. Postglacial distribution and species substructures: lessons from pollen, insects and hybrid zones. In Evolutionary patterns and processes. D. R. Lees and D. Edwards (Eds.) Academic Press, London, UK. pp. 97-123. Hewitt, G. 2000. The genetic legacy of the Quaternary ice ages. Nature 405: 907-913. Holsinger, K. E. 1991. Mass action models for plant mating systems: the evolutionary stability of mixed mating systems. American Naturalist 138: 606-622. Holsinger, K. E. 1996. Pollination biology and the evolution of mating systems in flowering plants. Evolutionary Biology 29: 107-149. Hu, X. - S., and R. A. Ennos. 1999. Scoring the mating systems of natural populations of three Larix taxa in China: L. gmelinii (Rupr.) Rupr., L. olgensis Henry and L. principis- rupprechtii Mayr. Scientia Silvae Sinicae 35: 21-31. 130 Husband, B.C., and D. W. Schemske. 1996. Evolution of the magnitude and timing of inbreeding depression in plants. Evolution. 50: 54-70. Innes, D. J., and G. G. Ringius. 1990. Mating system and genetic structure of two populations of white spruce (Picea glauca) in eastern Newfoundland. Canadian Journal of Botany 68: 1661-1666. Jaquish, B., and Y. A. El-Kassaby. 1998. Genetic variation of western larch in British Columbia and its conservation. Journal of Heredity 89: 248-253. Jarne, P., and D. Charlesworth. 1993. The evolution of selfing rate in functionally hermaphrodite plants and animals. Annual Review in Ecology and Systematics 24: 441-466. Johnston, M. O. 1998. Evolution of intermediate selfing rates in plants: pollination ecology versus deleterious mutations. Genetica 102/103: 267-278. Karhu, A., J. - H. Dieterich, and O. Savolainen. 2000. Rapid expansion of microsatellite sequences in pines. Molecular Biology and Evolution 17: 259-265. Karkkainen, K., and O. Savolainen. 1993. The degree of early inbreeding depression determines the selfing rate at the seed stage: model and results for Pinus sylvestris (Scots pine). Heredity 71: 160-166. Kijas J. M. H, and J. C. S.Fowler.1994. Enrichment of microsatellites from the citrus genome using biotinylated oligonucleotide sequences bound to streptavidin-coated magnetic particles. Biotechniques 16: 657-662. Kimura, M., and J. F. Crow. 1964. The number of alleles that can be maintained in a finite population. Genetics 49: 725-738. Kirkpatrick, M., and P. Jarne. 2000. The effects of a bottleneck on inbreeding depression and the genetic load. The American Naturalist 155: 154-167. Kjaer, E. D., and V. Suangtho. 1995. Outcrossing rate of teak (Tectona grandis (L.)). Silvae Genetica 44: 175-177. Knowles, P., G. R. Furnier, M. A. Aleksiuk, and D. J. Perry, 1987. Significant levels of self- fertilization in natural populations of tamarack. Canadian Journal of Botany 65: 1087- 1091. Kormut'ak, A., and D. Lindgren. 1996. Mating system and empty seeds in silver fir (Abies alba Mill.). Forest Genetics 3: 231-235. Korn, R. W. 2001. Analysis of shoot apical organization in six species of the Cupressaceae based on chimeric behavior. American Journal of Botany 88: 1945-1952. Korol, L., N. Kara, K. Isik, and G. Schiller. 1997. Genetic differentiation among and within natural and planted Cupressus sempervirens L. eastern Mediterraneaen populations. Silvae Genetica 46: 151-155. Krakowski, J. 2001. Conservation genetics of whitebark pine (Pinus albicaulis Engelm.) in British Columbia. MSc, University of British Columbia. 131 Kuittinen, H., O. Muona, K. Karkainen, and Z. Borzan. 1991. Serbian spruce, a narrow endemic, contains much genetic variation. Canadian Journal of Forest Research 21: 363-367. Kuittinen, H., and O. Savolainen. 1992. Picea omorika is a self-fertile but outcrossing conifer. Heredity 68: 183-187. Kuser, J. E., T. R. Meagher, D. L. Sheely, and A. White. 1997. Allozyme frequencies in New Jersey and North Carolina populations of Atlantic white-cedar, Chamaecyparis thyoides (Cupressaceae). American Journal of Botany 84: 1536-1541. Lagercrantz, TJ., and N. Ryman. 1990. Genetic structure of Norway spruce (Picea abies): concordance of morphological and allozymic variation. Evolution 44: 38-53. Lai, H. - L., and M. - X. Wang. 1997. Study on mating system in man-made populations of Masson pine. Scientia Silvae Sinicae 33: 219-224. Lamy, S., A. Bouchard, and J. - P. Simon. 1999. Genetic structure, variability, and mating system in eastern white cedar (Thuja occidentalis) populations of recent origin in an agricultural landscape in southern Quebec. Canadian Journal of Forest Research 29: 1383-1392. Lande, R., and D. W. Schemske. 1985. The evolution of self-fertilization and inbreeding depression in plants. I. Genetic models. Evolution 39: 24-40. Lande, R., D. W. Schemske, and S. T. Schultz. 1994. High inbreeding depression, selective interference among loci, and the threshold selfing rate for purging recessive lethal mutations. Evolution 48: 965-978. Latta, R. G. 1995. The effects of embryo competition with mixed mating on the genetic load in plants. Heredity 75: 637-643. Latta, R. G., and J. B. Mitton. 1999. Historical separation and present gene flow through a zone of secondary contact in ponderosa pine. Evolution 53: 769-776. Ledig, F. T., B. Bermejo-Velazquez, P. D. Hodgskiss, D. R. Johnson, C. Flores-Lopez, and V. Jacob-Cervantes. 2000. The mating system and genie diversity in Martinez spruce, an extremely rare endemic of Mexico's Sierra Madre Oriental: an example of facultative selfing and survival in interglacial refugia. Canadian Journal of Forest Research 30: 1156-1164. Ledig, F. T., M. A. Capo-Arteaga, P. D. Hodgskiss, H. Sbay, S. Flores-Lopez, M. T. Conkle, and B. Bermejo-Velazquez. 2001. Genetic diversity and the mating system of a rare Mexican pinon, Pinus pinceana, and a comparison with Pinus maximartinezii (Pinaceae). American Journal of Botany 88: 1977-1987. Ledig, F. T., and M. T. Conkle. 1983. Gene diversity and genetic structure in a narrow endemic, Torrey pine (Pinus torreyana Parry ex Carr.). Evolution 37: 79-85. Ledig, F. T., M. T. Conkle, B. Bermejo-Velazquez, T. Eguiluz-Piedra, P. D. Hodgskiss, D. R. Johnson, and W. S. Dvorak. 1999. Evidence for an extreme bottleneck in a rare Mexican 132 pinyon: genetic diversity, disequilibrium, and the mating system in Pinus maximartinezii. Evolution 53: 91-99. Ledig, F. T., V. Jacob-Cervantes, P. D. Hodgskiss, and T. Eguiluz-Piedra. 1997. Recent evolution and divergence among populations of a rare mexican endemic, Chihuahua spruce, following holocene climatic warming. Evolution 51: 1815-1827. Levinson, G., and G. A. Gutman. 1987. Slipped-strand misparing: a major mechanism for DNA sequence evolution. Molecular Biology and Evolution 4: 203-211. Lewandowski, A., and J. Burczyk. 2000. Mating system and genetic diversity in natural populations of European larch {Larix decidua) and stone pine {Pinus cembrd) located at higher elevations. Silvae Genetica 49: 158-161. Lewandowski, A., J. Burczyk, and L. Mejnartowicz. 1991. Genetic structure and the mating system in an old stand of Polish larch. Silvae Genetica 40: 75-79. Li, P., and W. T. Adams. 1989. Range-wide patterns of allozyme variation in Douglas-fir {Pseudostuga menziesii). Canadian Journal of Forest Research 19: 149-161. Lian, C, M. Miwa, and T. Hegetsu. 2001. Outcrossing and paternity analysis of Pinus densiflora (Japanese red pine) by microsatellite polymorphism. Heredity 87: 88-98. Lin, T. - P., T. - Y. Lee, L. - F. Yang, Y. - L. Chung, and J. - C. Yang. 1994. Comparison of the allozyme diversity in several populations of Chamaecyparis formosensis and Chamaecyparis taiwanensis. Canadian Journal of Forest Research 24: 2128-2134. Lin, T. - P., C. - T. Wang, and J. - C. Yang. 1998. Comparison of genetic diversity between Cunninghamia konishii and C. lanceolata. Journal of Heredity 89: 370-373. Longauer, R., P. Zhelev, L. Paule, and D. Gomory. 1992. The mating system outcrossing rate and genetic differentiation in natural Scots pine {Pinus sylvestris L.) populations from Bulgaria. Biologia 47: 539-547. Loveless, M. D., J. L. Hamrick, and R. B. Foster. 1998. Population structure and mating system in Tachigali versicolor, a monocarpic neotropical tree. Heredity 81: 134-143. Luikart, G., F. W. Allendorf, J.-M. Cornuet, and W. B. Sherwin. 1998. Distortion of allele frequency distributions provides a test for recent population bottlenecks. The Journal of Heredity 89: 238-247. Malvoti, M. E., S. Fineschi, M. Morgante, and G. G. Vendramin. 1995. Mating system in naturalized Juglans regia L. population in Italy. In Population genetics and genetic conservation of forest trees. P. Baradat, W. T. Adams, and G. Muller-Starck (Eds.) Papers presented at an international symposium organized by IUFRO, held 24-28 August 1992 at Carcan-Maubuisson, France. SPB Academic Publishing. Amsterdam, Netherlands, pp. 305-308. Mandal, A. K., and R. A. Ennos. 1995. Mating system analysis in a natural population of Acacia milotica subspecies kraussiana. Forest Ecology and Management 79: 235-240. 133 Marshall, H. D., C. Newton, and K. Ritland. 2002. Chloroplast phylogeography and evolution of highly polymorphic microsatellites in lodgepole pine (Pinus contorta). Theoretical and Applied Genetics 104: 367-378. Maruyama, T., and P. A. Fuerst. 1984. Population bottlenecks and nonequilibrium models in population genetics. I. Allele numbers when populations evolve from zero variability. Genetics 108: 745-763. Maruyama, T., and P. A. Fuerst. 1985. Population bottlenecks and nonequilibrium models in population genetics. II. Number of alleles in a small population that was formed by a recent bottleneck. Genetics 111: 675-689. Matheson, A. C., J. C. Bell, and R. D. Barnes. 1989. Breeding systems and genetics in some Central American pine populations. Silvae Genetica 38: 107-113. Matthes-Sears, U., S. C. Stewart, and D. W. Larson. 1991. Sources of allozymic variation in Thuja occidentalis in southern Ontario, Canada. Silvae Genetica 40: 100-105. McGranahan, M., J. C. Bell, G. F. Moran, and M. Slee. 1997. High genetic divergence between geographic regions in the highly outcrossing species Acacia aulacocarpa (Cunn. ex Benth.). Forest Genetics 4: 1-13. Mehringer, P. J. Jr. 1985. Late-Quaternary pollen records from the interior Pacific Northwest and northern Great Basin of the United States. In Pollen Records of Late-Quartenary North American Sediments. J. V. Bryant and R. G. Holloway (Eds.) American Association of Stratigraphic Palynologists Foundation. Dallas, Texas, pp. 167-189. Mejnartowicz, L., and A. Lewandowski. 1994. Allozyme polymorphism in seeds collected from a IUFRO-68 Douglas-fir test-plantation. Silvae Genetica 43: 181-186. Menotti-Raymond, M., and S. J. O'Brien. 1993. Dating the genetic bottleneck of the African cheetah. Proceedings of the National Academy of Science 90: 3172-3176. Merzeau, D., B. Comps, B. Thiebaut, and J. Letouzey. 1994. Estimation of Fagus sylvatica L. mating system parameters in natural populations. Annales des Sciences Forestieres 51: 163-173. Millar, C. I., and K. A. Marshall. 1991. Allozyme variation of Port-Orford-Cedar (Chamaecyparis lawsoniana): implications for genetic conservation. Forest Science 37: 1060-1077. Millar, M. A., M. Byrne, D. J. Coates, M. J. C. Stukely, and J. A. McComb. 2000. Mating system studies in jarrah, Eucalyptus marginata (Myrtacaeae). Australian Journal of Botany 48: 475-479. Minore, D. 1983. Western redcedar: A literature review. Pacific Northwest Forest and Range Experiment Station. USDA Forest Service General Technical Report PNW-150. Portland, Oregon. 70 pps. 134 Minore, D. 1990. Western redcedar (Thuja plicata Donn ex D. Don). In Silvics of North America, Vol. 1, Conifers. R. M. Bums and B. H. Honkala (Eds.) USDA Forest Service Agriculture Handbook 654 Washington, DC. pp. 590-600. Mitton, J. B. 1992. The dynamic mating systems of conifers. New Forests 6: 197-216. Mitton, J. B., Y. B. Linhart, M. L. Davis, and K. B. Sturgeon. 1981. Estimation of outcrossing in ponderosa pine, Pinus ponderosa Laws, from patterns of segregation of protein polymorphisms and from frequencies of albino seedlings. Silvae Genetica 30: 117-121. Mitton, J. B., R. G. Latta, and G. E. Rehfeldt. 1997. The pattern of inbreeding in Washoe pine and survival of inbred progeny under optimal environmental conditions. Silvae Genetica 46:215-219. Mitton, J. B., Y. B. Linhart, K. B. Sturgeon, and J. L. Hamrick. 1979. Allozyme polymorphisms detected in mature needle tissue of ponderosa pine. Journal of Heredity 70: 86-89. Moran, G. F., and W. T. Adams. 1989. Microgeographical patterns of allozyme differentiation in Douglas-fir from southwest Oregon. Forest Science 35: 3-15. Moran, G. F., J. C. Bell, and A. C. Matheson. 1980. The genetic structure and levels of inbreeding in a Pinus radiata D. Don seed orchard. Silvae Genetica 29: 190-193. Moran, G. F., J. C. Bell, and K. G. Eldridge. 1988. The genetic structure and the conservation of the five natural populations of Pinus radiata. Canadian Journal of Forest Research 18: 506-514. Moran, G. F., P. Muona, and J. C. Bell. 1989. Breeding systems and genetic diversity in Acacia auriculiformis and A. crassicarpa. Biotropica 21: 250-256. Morgan, M. T., D. J. Schoen and T. M. Bataillon. 1997. The evolution of self-fertilization in perennials. The American Naturalist 150: 618-638. Morgan, M. T. 2001. Consequences of life history for inbreeding depression and mating system evolution in plants. Proceedings of the Royal Society of London: Series B 1817-1824. Morgante, M., P. Rossi, G. G. Vendramin, and G. Boscherini. 1994. Low levels of outcrossing in Pinus leucodermis: further evidence in artificial stands. Canadian Journal of Botany 72: 1289-1293. Morgante, M., G. G. Vendramin, and A. M. Olivieri. 1991b. Mating system analysis in Pinus leucodermis Ant.: detection of self-fertilization in natural populations. Heredity 67: 197- 203. Morgante, M., G. G. Vendramin, and P. Rossi. 1991a. Effects of stand density on outcrossing rate in two Norway spruce (Picea abies) populations. Canadian Journal of Botany 69: 2704-2708. Morgante, M., G. G. Vendramin, P. Rossi, and A. M. Olivieri. 1993. Selection against inbreds in early life-cycle phases in Pinus leucodermis Ant. Heredity 70: 622-627. 135 Mosseler, A., D. J. Innes, and B. A. Roberts. 1991. Lack of allozyme variation in disjunct Newfoundland populations of red pine (Pinus resinosa). Canadian Journal of Forest Research 21: 525-528. Muona, O. 1988. Population genetics in forest tree improvement. In Plant population genetics, breeding and genetic resources. A. D. H. Brown, M. T. Clegg, A. L. Kahler, and B. S. Weir (Eds.) Sinauer Associates Inc., Sunderland, Massachusetts, pp. 282-298. Muona, O., and A. Hariu. 1989. Effective population sizes, genetic variability, and mating system in natural stands and seed orchards of Pinus sylvestris. Silvae Genetica 38: 221- 228. Murawski, D. A., B. Dayanandan, and K. S. Bawa. 1994a. Outcrossing rates of two endemic Shorea species from Sri Lankan tropical rain forests. Biotropica 26: 23-29. Murawski, D. A., I. A. U. N. Gunatilleke, and K. S. Bawa. 1994b. The effects of selective logging on inbreeding in Shorea megistophylla (Dipterocarpaceae) from Sri Lanka. Conservation Biology 8: 997-1002. Murawski, D. A., J. L. Hamrick, S. P. Hubbell, and R. B. Foster. 1990. Mating systems of two Bombaceous trees of a neotropical moist forest. Oecologia 82: 501-506. Murawski, D. A., and J. L. Hamrick. 1991. The effect of the density of flowering individuals on the mating systems of nine tropical tree species. Heredity 67: 167-174. Nagamitsu, T., S. Ichikawa, M. Ozawa, R. Shimamura, N. Kachi, Y. Tsumura, and N. Muhammad. 2001. Microsatellite analysis of the breeding system and seed dispersal in Shorea leprosula (Dipterocarpaceae). International Journal of Plant Science 162: 155- 159. Nakamura, R. R., and N. C. Wheeler. 1992. Self-fertility variation and paternal success through outcrossing in Douglas fir. Theoretical and Applied Genetics 83: 851-854. Namkoong, G., and J. Bishir. 1987. The frequency of lethal alleles in forest tree populations. Evolution 41: 1123-1126. Neale, D. B., and W. T. Adams. 1985. Allozyme and mating-system variation in balsam fir (Abies balsamea) across a continuous elevational transect. Canadian Journal of Botany 63: 2448-2453. Nei, M. 1972. Genetic distance between populations. American Naturalist 106: 283-292. Nei, M., T. Maruyama, and R. Chakraborty. 1975. The bottleneck effect and genetic variability in populations. Evolution 29: 1-10. Newcombe, R. G. 1998. Two-sided confidence intervals for the single proportion: comparison of seven methods. Statistics in Medicine 17: 857-872. O'Connell, L. M., and C. E. Ritland. 2000. Characterization of microsatellite loci in western redcedar (Thuja plicata). Molecular Ecology 9: 1920-1922. 136 O'Connell, L. M., F. Viard, J. Russell, and K. Ritland. 2001. The mating system in natural populations of western redcedar (Thuja plicata). Canadian Journal of Botany 79: 753- 756. Oetting, W. S., H. K. Lee, D. J. Flanders, G. L. Wiesner, T. A. Sellers, and R. A. King. 1995. Linkage analysis with multiplexed short tandem repeat polymorphisms using infrared fluorescence and M13 tailed primers. Genomics 30: 450-458. Ohta, T., and M. Kimura. 1973. A model of mutation appropriate to estimate the number of electrophoretically detectable alleles in a finite population. Genetic Research Cambridge 22: 201-204. O'Malley, D. M., and K. S. Bawa. 1987. Mating system of a tropical rain forest tree species. American Journal of Botany 74: 1143-1149. O'Malley, D. M., D. P. Buckley, G. T. Prance, and K. S. Bawa. 1988. Genetics of Brazil nut (Bertholletia excelsa Humb. & Bonpl.: Lecythidaceae), 2. Mating system. Theoretical and Applied Genetics 76: 929-932. Omi, S.K., and W. T. Adams, 1986. Variation in seed set and proportions of outcrossed progeny with clones, crown position, and top pruning in a Douglas-fir seed orchard. Canadian Journal of Forest Research 16: 502-507. Orive, M. E. 2001. Somatic mutations in organisms with complex life histories. Theoretical Population Biology 59: 235-249. Otto, S. P., and I. M. Hastings. 1998. Mutation and selection within the individual. Genetica 102/103: 507-524. Owens, J. N., and M. Molder. 1980. Sexual reproduction in western red cedar (Thuja plicata). Canadian Journal of Botany 58: 1376-1393. Owens, J. N., A. M. Colangeli, and S. J. Morris. 1990. The effect of self-, cross-, and no pollination on ovule, embryo, seed, and cone development in western red cedar (Thuja plicata). Canadian Journal of Forest Research 20: 66-75. Page, R. D. M. 1998. Tree View for Macintosh. Version 1.5. University of Glasgow, UK. Paiva, J. R. D., P. Y. Kageyama, and R. Vencovsky. 1993. Outcrossing rates and inbreeding coefficients in rubber trees (Hevea brasiliensis (Willd. ex Adr. de Juss.) Mueller. Arg.). Revista Brasiliera de Genetica 16: 1003-1011. Parducci, L., A. E. Szmidt, F. Villani, X.-R. Wang, and M. Cherubini. 1996. Genetic variation of Abies alba in Italy. Hereditas (Landskrona) 125: 11-18. Park, W. S., and D. P. Fowler. 1984. Inbreeding in black spruce (Picea mariana (Mill.) B.S.P.): self-fertility, genetic load, and performance. Canadian Journal of Forestry Research 14: 17-21. Parker, T., and F. D. Johnson, 1988. Seed and vegetative regeneration of western redcedar in the northern Rocky Mountains. In Western red cedar-does it have a future? N. J. Smith (Ed.) 137 University of British Columbia, Faculty of Forestry, Vancouver, British Columbia, pp. 136-144. Paule, L., D. Lindgren, and R. Yazdani. 1993. Allozyme frequencies, outcrossing rate and pollen contamination in Picea abies orchards. Scandinavian Journal of Forest Research 8: 8-17. Perez-Nasser, N., L. E. Eguiarte, and D. Pinero. 1993. Mating system and genetic structure of the distylous tropical tree Psychotriafaxlucens (Rubiaceae). American Journal of Botany 80: 45-52. Perry, D. J., and B. P. Dancik. 1986. Mating system dynamics of Lodgepole Pine in Alberta, Canada. Silvae Genetica 35: 190-195. Perry, D. J., and P. Knowles, 1990. Evidence of high self-fertilization in natural populations of eastern white cedar (Thuja occidentalis). Canadian Journal of Botany 68: 663-668. Perry, D. J., P. Knowles, and F. C. Yeh, 1990. Allozyme variation of Thuja occidentalis L. in Northwestern Ontario. Biochemical Systematics and Ecology 18: 111-115. Politov, D. V., and K. V. Krutovskii. 1990. Genetic variation in Pinus sibirica du Tour. V. Analysis of mating system. Genetika 26: 2002-2011. Potenko, V. V., and A. V. Velikov. 1998. Genetic diversity and differentiation of natural populations of Pinus koraiensis (SIEB. et Zucc) in Russia. Silvae Genetica 47: 202-208. Potenko, V. V., and A. V. Velikov. 2001. Allozyme variation and mating system of coastal populations of Pinus koraiensis Sieb. et Zucc. in Russia. Silvae Genetica 50: 3-4. Primmer, C. R., H. Ellegren, N. Saino, and A. P. M0ller. 1996. Directional evolution in germline microsatellite mutations. Nature Genetics 391-393. Puglisi, S., R. Lovreglio, and M. Attolico. 1999. Subpopulation differentiation along elevational transects within two Italian populations of Scots pine (Pinus sylvestris L.). Forest Genetics 6: 247-256. Raddi, S., and S. Sumer. 1999. Genetic diversity in natural Cupressus sempervirens L. populations in Turkey. Biochemical and Systematic Ecology 27: 799-814. Rajora, O. P., A. Mosseler, and J. E. Major. 2000. Indicators of population viability in red spruce, Picea rubens. II. Genetic diversity, population structure, and mating behavior. Canadian Journal of Botany. 78: 941-956. Raymond, M., and F. Rousset. 1995. GENEPOP (version 1.2) population genetics software for exact tests and ecumenicism. Journal of Heredity 86: 248-249. Rehfeldt, G.E., 1994. Genetic structure of western red cedar populations in the interior west. Canadian Journal of Forest Research 24: 670-680. Reusch, T. B. H. 2000. Pollination in the marine realm: microsatellites reveal high outcrossing rates and multiple paternity in eelgrass Zostera marina. Heredity 85: 459-464. Reusch, T. B. H. 2001. Fitness-consequences of geitonogamous selfing in a clonal marine angiosperm (Zostera marina). Journal of Evolutionary Biology 14: 129-138. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43: 223-225. 138 Richards, A. J. 1986. Plant Breeding Systems. George Allen & Unwin. London, U.K. 529 p. Richardson, B. A., S. J. Brunsfeld, and N. B. Klopfenstein. 2002. DNA from bird-dispersed seed and wind-disseminated pollen provides insights into postglacial colonization and population genetic structure of whitebark pine (Pinus albicaulis). Molecular Ecology 11: 215-227. Ritland, K. 1983. Estimation of mating systems. In Isozyme in plant genetics and breeding, Part A. S.D. Tanksley and TJ. Orton (Eds.). Elsevier Scientific Publications, Amsterdam, pp. 289-302. Ritland, K. 1989. Correlated matings in the partial selfer, Mimulus guttatus. Evolution, 43: 848- 859. . • , • . • Ritland, K. 1990a. A series of FORTRAN computer programs for estimating plant mating systems. Journal of Heredity 81: 235-237. Ritland, K. 1990b. Inferences about inbreeding depression based on changes of the inbreeding coefficient. Evolution 44: 1230-1241. Ritland, K. 2002. Estimation of gene frequency and heterozygosity from pooled samples. Molecular Ecology Notes 2: 370-372. Ritland, K., and Y. A. El-Kassaby. 1985. The nature of inbreeding in a seed orchard of Douglas fir as shown by an efficient multilocus model. Theoretical and Applied Genetics 71: 275- 384. Ritland, C, T. Pape, and K. Ritland. 2001. Genetic structure of yellow cedar (Chamaecyparis nootkatensis). Canadian Journal of Botany 79: 822-828. Rossi, P., G. G. Vendramin, and R. Giannini. 1996. Estimation of mating system parameters in two Italian natural populations of Fagus sylvatica. Canadian Journal of Forest Research 26: 1187-1192. Riiter, B., J. L. Hamrick, and B. W. Wood. 2000. Outcrossing rates and relatedness estimates in pecan (Carya illinoinensis) populations. The Journal of Heredity 91: 72-75. Sampson, J. F., S. D. Hopper, and S. H. James. 1989. The mating system and population genetic structure in a bird-pollinated mallee, Eucalyptus rhodantha. Heredity 63: 383-393. SAS Institute. 1997. JMP user's guide, version 3.2.1. SAS Institude Inc. Cary, North Carolina. Savolainen, O. K., Karkkainen and H. Kuittinen. 1992. Estimating numbers of embryonic lethals in conifers. Heredity 69: 308-314. Schlotterer, C, R. Ritter, B. Harr, and G. Brem. 1998. High mutation rate of a long microsatellite allele in Drosophila melanogaster provides evidence for allele-specific mutation rates. Molecular Biology and Evolution 15: 1269-1274. Schneider, S., D. Roessli, and L. Excoffier. 2000. Arlequin ver 2.000: A software for population genetics data analysis. Genetics and Biometry Laboratory, Dept. of Anthropology and Ecology, Geneva, Switzerland. URL: http://anthro.unige.ch/arlequin. 139 Schroeder, S. 1989. Outcrossing rates and seed characteristics in damaged natural populations of Abies alba Mill. Silvae Genetica 38: 185-189. Seavey, S. R., and K. S. Bawa. 1986. Late-acting self-incompatibility in angiosperms. The Botanical Review 52: 195-219. Shaw, D. V., and R. W. Allard. 1981. Analysis of mating system parameters and population structure in Douglas-fir using single-locus and multilocus methods. In Proceedings of symposium on isozymes of North American forest trees and insects, July 27, 1979. Pacific Southwest Forest and Range Exp. Stn., Forest Service, U.S. Department of Agriculture, Berkeley, Calififornia. pp. 18-22. Shaw, D. V., and R. W. Allard. 1982. Estimation of outcrossing rates in Douglas-fir using isozyme markers. Theoretical and Applied Genetics 62: 113-120. Shea, K. L. 1987. Effects of population structure and cone production on outcrossing rates in Engelmann spruce and subalpine fir. Evolution 41: 124-136. Shea, K. L. 1990. Genetic variation between and within populations of Engelmann spruce and subalpine fir. Genome 33: 1-8. Siegismund, H. R., E. D. Kjaer, and U. B. Nielsen. 1996. Mating system estimates and effective population numbers for an isolated noble fir (Abies procera) clonal seed orchard in Denmark. Canadian Journal of Forest Research 26: 1135-1141. Siegismund, H. R., and E. D. Kjaer. 1997. Outcrossing rates in two stands of noble fir (Abies procera Rehd.) in Denmark. Silvae Genetical44-146. Simon, J. P., Y. Bergeron, and D. Gagnon. 1986. Isozyme uniformity in populations of red pine (Pinus resinosa) in the Abitibi Region, Quebec. Canadian Journal of Forest Research 16: 1133-1135. Snyder, T. P., D. A. Stewart, and A. F. Strickler. 1985. Temporal analysis of breeding structure in jack pine (Pinus banksiana Lamb.). Canadian Journal of Forest Research 15: 1159- 1166. Soltis D. E., M. A. Gitzendanner, D. D. Strengem and P. S. Soltis. 1997. Chloroplast DNA intraspecific phylogeography of plants from the Pacific Northwest of North America. Plant Systematic and Evolution 206: 353-373. Sorensen, F. 1969. Embryonic genetic load in coastal Douglas-fir Pseudotsuga menziesii var. menziesii. The American Naturalist 103: 389-398. Sorensen, F. C. 1982. The roles of polyembryony and embryo viability in the genetic system of conifers. Evolution 36: 725-733. Sorensen, F. C. 1999. Relationship between self-fertility, allocation of growth, and inbreeding depression in three coniferous species. Evolution 53: 417-425. Sproule, A. T., and B. P. Dancik. 1996. The mating system of black spruce in north-central Alberta, Canada. Silvae Genetica 45: 159-164. 140 Stacy, E. A., J. L. Hamrick, J. D. Nason, S. P. Hubbell, R. B. Foster, and R. Condit. 1996. Pollen dispersal in low-density populations of three neotropical tree species. The American Naturalist 148: 275-298. Stauffer, A., and W. T. Adams. 1993. Allozyme variation and mating system of three Douglas-fir stands in Switzerland. Silvae Genetica 42: 254-258. Steinhoff, R. J., D. G. Joyce, and L. Fins. 1983. Isozyme variation in Pinus monticola. Canadian Journal of Forest Research 13: 1122-1132. Stern, K., and L. Roche. 1974. Genetics of Forest Ecosystems. Berlin: Springer-Verlag. Stebbins, G. L. 1950. Variation and evolution in plants. Columbia University Press, New York. 643 p. Stoehr, M. U., and Y. A. El-Kassaby. 1997. Levels of genetic diversity at different stages of the domestication cycle on interior spruce in British Columbia. Theoretical and Applied Genetics 94: 83-90. Strauss, S. H., and M. T. Conkle. 1986. Segregation, linkage, and diversity of allozymes in knobcone pine. Theoretical and Applied Genetics 72: 483-493. Tremblay, M., and J. P. Simon. 1989. Genetic structure of marginal populations of white spruce (Picea glauca) at its northern limit of distribution in Nouveau-Quebec. Canadian Journal of Forest Research 19: 1371-1379. Truesdale, H. D., and L. R. McClenaghan. 1998. Population genetic structure of tecate cypress (Cupressaceae). Southwestern Naturalist 43: 363-373. Turgeon, J., and L. Bernatchez. 2001. Clinal variation at microsatellite loci reveals historical secondary intergradation between glacial races of Coregonus artedi (Teleostei: coregoninae). Evolution 11: 2274-2286. Tuskan, G. A., K. E. Francis, S. L. Russ, W. H. Romme, and M. G. Turner. 1996. RAPD markers reveal diversity within and among clonal and seedling stands of aspen in Yellowstone National Park, USA. Canadian Journal of Forest Research 26: 2088-2098. Uchida, K., N. Tomaru, C. Tomaru, C. Yamamoto, and K. Ohba. 1997. Allozyme variation in natural populations of Hinoki, Chamaecyparis obtusa (Sieb. Et Zucc.) Endl and its comparison with the plus-tree selected from artificial stands. Breeding Science 47: 7-14. Udupa, S. M., and M. Baum. 2001. High mutation rate and mutational bias at (TAA)n microsatellite loci in chickpea (Cicer arietum L.). Molecular Genetics and Genomics 265: 1097-1103. Uyenoyama, M. K., K. E. Holsinger, and D. M. Waller. 1993. Ecological and genetic factors directing the evolution of self-fertilization. Oxford Surveys of Evolutionary Biology 9: 327-381. Vazquez, J. F., T. Perez, J. Albornoz, and A. Domfnguez. 2000. Estimation of microsatellite mutation rates in Drosophila melanogaster Genetical Research 76: 323-326. 141 Vidakovic, M. 1991. Conifers: morphology and variation (M. Soljan, Translator). Graficki zavod Hrvatske, Zagreb. 756 p. Vigouroux, Y., J. S. Jaqueth, Y. Matsuoka, O. S. Smith, W. D. Beavis, J. S. C. Smith, and J. Doebley. 2002. Rate and pattern of mutation at microsatellite loci in maize. Molecular Biology and Evolution 19: 1251-1260. vonRudloff, E., and M. S. Lapp. 1979. Populational vairation in leaf oil terpene composition of western red cedar, Thuja plicata. Canadian Journal of Botany 57: 476-479. vonRudloff, E., M. S. Lapp, and F. C. Yeh. 1988. Chemosystematic study of Thuja plicata: multivariate analysis of leaf oil terpene composition. Biochemical Systematics and Ecology 16: 119-125. Wang, C. - T., and T. - P. Lin. 1998. Inheritance and linkage relationships of allozymes and estimation of outcrossing rates in a seed orchard of Cunninghamia konishii Hay. Silvae Genetica 47: 33-37. Warner, B. G., and R. W. Mathewes. 1982. Ice-free conditions on the Queen Charlotte Islands, British Columbia, at the height of late Wisconsin glaciation. Science 218: 667. Weir, B. S. 1990. Genetic Data Analysis. Sinauer. Sunderland, U.S.A Weir, B. S., and C. C. Cockerham. 1984. Estimating F-statistics for the analysis of population structure. Evolution, 38: 1358-1370. Wheeler, N. C, and R. P. Guries. 1982. Population structure, genie diversity, and morphological variation in Pinus contorta Dougl. Canadian Journal of Forest Research 12: 595-606. Whitham, T. G., and C. N. Slobodchikoff. 1981. Evolution by individuals, plant-herbivore interactions, and mosaics of genetic variability: The adaptive significance of somatic mutation in plants. Oecologia 49: 287-292. Williams, C. G., Y. Zhou, and S. E. Hall. 2001. A chromosomal region promoting outcrossing in a conifer. Genetics 159: 1283-1289. Wilson, E. B. 1927. Probable inference, the law of succession, and statistical inference. Journal of the American Statistical Association 22: 209-212. Xie, C. Y., B. P. Dancik, and F. C. Yeh. 1991. The mating system in natural populations of Thuja orientalis. Canadian Journal of Forest Research 21: 333-339. Xie, C. Y., B. P. Dancik, and F. C. Yeh. 1992. Genetic Structure of Thuja orientalis. Biochemical Systematics and Ecology 20: 433-441. Xie, C. Y., and P. Knowles. 1994. Mating system and effective pollen immigration in a Norway spruce (Picea abies (L.) Karst) plantation. Silvae Genetica 43: 48-52. Yeh, F. C. 1988. Isozyme variation of Thuja plicata (Cupressaceae) in British Columbia. Biochemical Systematics and Ecology 16: 373-377. Yeh, F. C, and Y. A. El-Kassaby. 1980. Enzyme variation in natural populations of Sitka spmce (Picea sitchensis). 1. Genetic variation patterns among trees from 10IUFRO provenances. Canadian Journal of Forest Research 10: 415-422. 142 Yeh, F. C, and D. O'Malley. 1980. Enzyme variations in natural populations of Douglas-fir, Pseudotsuga menziesii (Mirb.) Franco, from British Columbia. 1. Genetic variation patterns in coastal populations. Silvae Genetica 29: 83-92. Zheng, Y. Q., and R. Ennos. 1997. Changes in the mating systems of populations of Pinus caribaea Morelet var. caribaea under domestication. Forest Genetics 4: 209-215. Zheng, Y. Q., and R. A. Ennos. 1999. Genetic variability and structure of natural and domesticated populations of Caribbean pine (Pinus caribaea Morelet). Theoretical and Applied Genetics 98: 765-771. 143 Appendix I Description of genetic diversity parameters PPL: Percentage of polymorphic loci, calculated as the number of loci with more than one allele divided by the total number of loci assayed. A/L: Number of alleles per locus averaged. Includes both polymorphic and monomorphic loci. Hes: Total genetic diversity for a species was obtained by averaging genetic diversity over all loci (polymorphic and monomorphic). where p. is the mean frequency of ith allele across all populations analyzed for a species. Hep: Total genetic diversity within a population and was obtained by averaging genetic diversity over all loci in a population (including polymorphic and monomorphic). th where p, is the frequency of i' allele in a population, and Hepis the mean over all populations. HT: Total genetic diversity for a species calculated using only polymorphic loci Hs: Mean diversity within a population calculated using only polymorphic loci Gsl: The proportion of the genetic diversity residing among populations of a species. Gst is the average over all polymorphic loci. Tt Tf ON a CJ > ON o o XI ON ON JD ON p- ON ON CJ >> ON 2 ON B 43 ON ON CJ cu O o ~ -s on ON cS e o 00 a ON CN « CO ON 43 CJ W *. • CO 04 ON o 3 « • N^ cS B PH CS WW. o O o "3 m PH PH I 4 o « I CJ o DH ON PH ON 3 43 OH S .2 CM -t—» op J3 u O -4—» * a- •a o 1 O CU CU a CS CU J3 if o J3 iJ5 U 43 o o o -35 c55 CN CN .5 >- o oo IT) oin oe n or - o oTt " O o — 60 CJ C5 c> d d d d d d N o CJ 3 cs a o cj CO CJ -4-1 43 — 00 r- NO CN CJ o m i> & NO o in r- ON CN NO >n NO B O 43 O CO NO r- CN CO NO NO 00 r~ XI 3 , , CJ XJ Tf CN m NO CO r- NO 43 CJ CJ o NO ON 00 CN 00 cn >n — CO PH CN CN CN CN CN >N X PH CJ cj " > B CJ cS CJ 'o CJ 60 ON CJ Tf 00 cn oo in m m s .2 CN ON ON 00 1 a a oo /1 —— CoN CJ' CJ o o o o d d d d d d en l-l CJ oo cj CJ OH CO > T(" 00 CO 00 NO CN CN CO CO ON 5 cS CN ON 00 00 NO i—nH OroN .—1 00 —— .—( m-H o -H XJ •a s — — — — O CN CN CO sf I&: o d d d d d d d d d d d d a •3 £ 0, a XJ 3 o o o < O 00 ON ON m ON CN ON TT NO 3 r- - co o CJ o cj 3 4= PH 4-* PH JD tt, m >n 00 O NO CN TI• c~ Tt 1> O O 1- CN CN CN CN CS CN CN m PH a =3 O J> XJ «» t '<-> 5 a _J3o VH S2 V » a "3 "c3 "c3 "ca "ca "c3 "ca "c3 la PH CJ •« S3 43 |3 ]3 3 43 o o o | % o PH 03 ca ca ca ca ca ta ca la 1- la la - CJ _a CS 3 C B B B B B B B B B B o B B COJ B >> CO B 'S3 43 CJ 2 1 ca ca ca ca ca ca ca ca ca ca ca ca ca ca 3 CU CU CD CU cu cu cu cu o cu cu cu cu X) -4—* o o o • o uo o o CJ o o CJ CJ o o C3 ca ca ca ca ca ca ca ca ca ca ca ca ca CJ a B B B B B B B B B B B B B B SH CU Cu CU CU CU Cu CU CU I O X! £ £ £ cu CO *CO CJ 60 XJ 4 O o-1 60 >> N 5t cu cu O 13 6 « es Ii s '> cu 0J. a a 3 CU U o 43 a o C3 a X) X) '3 c/3 8 3 CU o 00 CJ X cj B ^5 00 00 3 -fcH 60 CJ R .a CS C3 a a c c 00 CJ Cj cu 00 H CU a CJ C3 a B <" PH fc cs cu CU X > ia Cj CU cs x) T- "5 TC 3 5 § CU a CN 00 ~. OS Os Os Os Os ^ 00 00 OS o Os Os Os 2 .S3 o o o o o o 2 o o Os o o B o o O o 3 o o ON CN CN O o o B CN CN CN CN S •a U « cy a W B CJ "0 CO o o 00 a. 60 3 — *j T3 T3 •3 .3 g 1 O O o cj o O m —1 OH C/5 N I CN SO © o o o o o 00 os c- cn r- •t m r- SO 0 0 CN OS so CO O CN OO 06 cn CO •>* "* t- so SO m >n 00 so •<* r- SD CN CN OS co r~ SO r-» m m •* m CO t-; co SO OS SO CN •ct 00 co SO CN CN Os 51 CN CN CN CN CN on a Os r- CO SO so OS ro m >n 00 m 179 5 0 CN .227 4 CN CN O 0 O 0 O 0 0 0 0 O 0 0 P- r~ r~ CN m 00 SO LZZ CO SO in - CO >n m 00 00 m •* 00 m 0 .246 5 co CN CN CN CN CN O 0 0 O 0 O •, 0 O 0 0 0 0, O 0 O O 0 0 0 PH O CN >n CN 0 O Os in CN CN in 00 in CH CN 00 00 co V~i CN Os r~ 00 SO SO CN O SO CN CN CN m CN CO CN T3 13 13 13 13 13 13 13 13 13 13 T3 13 13 13 J2 b b b b b b 3 13 S3 P bP P P P p % td ca cd J3 P -cad -c5d cd B B B cBd Bcd cBd B C B B B CJ CJ u CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ E C3 cCdJ cCdJ cd cCdJ cCdJ CJ cCdJ cCdJ cCdJ cd cd cCsJ cd cd cCdJ cCdJ cCdJ cCdJ cCdJ 0 0 c0 j 0 0 0 0 cCJj cj 0 c0 j u0 0 CJ 0 0 CJ cd cd cd cd cdj OCO cd cd cd cd cud cd cd cd cd cd cd cd cd fi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n Pi n T3 cu 3 .S B o o a a c c a a O a a. a cu CN oOso CN 2 00 00 ?J 00 00 > Os Os Os OS OS OS OS LL. Os C i— — Os <& o Os o ON B o 60 2 >, O •3 "S B CJ CJ O CN o "cj 60 CJ X K <3 C —1 b > is o JS33 ca ca 60 60 =3 O <*) •S « -s B B B 60 CJ •3 O o o o CJ o 3 3 Jc«i O -J u CO J3 O o 03 Si s SO CN m CN CO CU o SO CN •* o m CO O o o o o o o O o CO m CO in -* SO oo 00 men O SCON SO SmO c~ msb SO O oo SO CO •<* in SO OS oo CO t-; so t-; Ov CN oo t-; so OS ~* CN " Os CN 00 00 CN so CN 00 00 SO -C*N OS SO so 00 00 SO average . in CN CN a CN o o in O o o o ©' o © O o d o o o o CO m SO SO CN 00 00 CN 00 SO SO CO SO CN •<* m 00 00 r- "* OS "3- 00 OS 00 m r- CN so CN o CN so CN las o © © ©' ©' ©' o o ©' ©' ©' o O o o CON o o o o oo CO - - m 00 CON m CN 00 CO CN CN PO P CN CO CO oo CSON CO SO •* CN SO CN CO o CN CN CN CN o "ca 13 13 13 13 13 13 13 13 13 13 13 13 13 1 13 13 13 13 % % ca ca In ca ca ca ca CJ ca s B ca f3a B ca ca ca B B B ca B B o ca ca ca c B B B B B B B B B S CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ ca cCaJ cCaJ cCaJ cCaJ CJ CJ CJ CJ CJ CJ CJ CJ CJ cj CJ CJ CJ C3 o cj cj CO cj cj cj o B sca ca ca Bca ca cBa cBa B Eca Bca cBa Bca Bca Bca cBa cBa Bca cBa 0- CU CM CL, OH OH OH OH OH OH OH CL, CU Cu Cu CU CU 0. OH T3 CD a a o o a a i: ^ <3 fi• a 00 t/3 S 1 c ts «o oj 3 3 o u 3 "3 §P to a Q >> c £ £ O ft. a to to to ? ts o •8 CD 3 3 3 3to 3to 3 3 3 3 3 3 3 3 "a 3 K C C C S K S S B S C cu Cu K C 3 S| SI £1 SI S| s: s: SI s: s; s: s; SI cu oo ON ON oNO\ ON aa •a ON ON I NO ON ON ON ON ON t- OO OO OO ON 13 o ON f- U o ON ON ON tu ON ON ON 2 o as ^i o Tt ON cj ON ON CN oS tO to co o ON 0O CN ON CO T3 T3 T3 =3 CS < < < S3 •o JJ „ =3 <=3 =8 CO 2 o C fi O o CU 5 CJ OH NO NO ON in ON Tt Tt m CO Tt CO o o O o o 8 d d d d d d d m CN Tt to CO Tt ON NO CN in 00 CN CN m oo CN Tt •*t CN m co CO Tt o m o m m oo CO m Tt 00 co 00 NO CN NO in CN Tt cn vi 00 m in oo Tt in r- CN oo 00 oo in CN ON NO Tt ON CN 00 f-; CO NO 1 in CO 5 ' CN CN 0J3 C3 tU >• CO CN CN 00 CN 00 CN m 00 00 Tt O CN oo O m CN m t> CN O a o o o d d d d d d d d d d d d in r- r> CO CN CN oo CN 00 CN ON NO oo 00 Tt O oo Tt in CN 3 o or- CN in CN CN o o m o CN l»5 d d d d d d d d d d d d d d d d d d OH TT ON ON m NO CN r» NO Tt CN NO Ti• Tt CO CN CO CN CN o co CN CH m ON CN o Tt o 00 CN CO o CtoN o O o r- CN CN CN CN CN CN CN 13 13 13 13 13 13 13 13 3 b b b b b b b3 -3 b-J 1 M CS 13 ca •a ea ca ca nat i B B B B B B B B ca B CU CU CU s CCS C3 tCuO tCCuS tu a o ON CN ON S3 ON ON C/3 CO cs C „i, CJ X ffl CO NO CO NO NS NO o e e ON o CO CO CN NO CN © Ct~O ltO- NO in ON CN oo -* ON CO ON NO Tf 3 iccT o o o O CN o o o o O O © © e PH 00 NO CN ON © os x; T3 NO CO S3 ir, CO *H II II HH D. cx 3 e s u 3 u u u S O id u CD =S .a a o CU cu cu o s en e/o so 2 cs a cs u u c s C3 CU cu cu ex s s s > >• > :s :§ < < T3 C/J cj cj <3 a © cj o a, a, o> o a e S OH 3- c < S cS 3 ON Tt x) S3 O X! O p ON o ON NO NO —t CN ON ON X) ON M ON ON CD ON >> CD N co Tt Tt ZLJ CJ °> ON -O as S3 Si ON ON a CO ON ON 2 « "c3 00 ON ON s I =3 =3 -S 3 •a o, ^ 3 ^ B •s -4—* 3 =3 o o < -a rt B 5 E £> 73 w O g 3 2 o =3 o z "O tu ~* •C^ "3 CO a 3 > w o) .3 3 W X C3 CU j3 o BC > Q Z M CN a CO CN r- CO CD CN 00 1- 1-1 oo B B oo ON ON C- CO NO ON m NO U~t ON NO ON 00 CN CPN- 00 ON CD CO 00 d d d 00 OO Tt CM d d d d d d o o l-H CJ 00 NO ON m ON ON Tt CN CO m NO ON Tt co Tt -4—> ON NO ON ON 3 B •a OTNf oo r~ 00 m OTNt Nr~O CN o CO O s d oo d d d 00 OO r» 00 00 d 00 x) O d d d d d d d d d CD CJ CD CD > NO o ON ON Tt oo CN CO UO CO oo CN CO CO CM ON ON ON ON OTNt o OO tr> OTNt NO o 1 OoN ON o Tt CO O © O d 00 CN d 00 d 00 I OO >A 4 d d d - d d d d CN NO NO 1 • • • 1 ^ IoD r-; OoN NO a en < Tt r- CO Tt xi NO oo ro C- o CN B d NO 00 ON 00 ft rt d O d d d 3 PH ON ' oo Tt CN CN ON u-> CN CN CN O a PH o o o CN CO oo CN CO ON 00 t> >r> r- NO o CJ to CD 73 73 73 73 B 13 13 13 13 • -H B S3 13 •3 S3 es a JB t J3 S3 o CLT a B CD a o o CM O o CO o OH CN Xl" CJ CJ CJ cu tu CJ tu CU CJ CU CJ CJ CJ CU CJ CU tu CJ > CCCJS CcsU tcus CcsJ CcsU CcsU tCuS CU3 CctJf CcSJ CcSJ CcsU tCuS tCuS CUS CSJ OO l-H CJ CJ CJ CJ CJ CJ CJ o CJ o o CJ CJ o o CJ B 1 Bcs cs Bcs Bcs cBs cBs Bcs BCO Bcs BcS BCS ICBS Btc ! BICS BcS BcS a C8 s CO I* a. 'o O o 33 O CD 4-1 O 1 3 cj a B s CD O s 8 l CJ Cj CJ cj C3 8 5 CJ CJ a. o o •3 -CS •s hJ PH en TC TC -a TC TC TC 3 3 3 3 3 3 TC CN ON CN ON ON ON NO ON o ON ON ON ON ON NO •a o ON 00 O o ON 3 CB J ON O o o 'So o O o c JS Q o CN ~ 00 2 « CO c^j s 1 >N cu ON Tg °^ — o N E? 2? \£3 w 5-" O I •a 3 T3 CJ ">> 7j >> o CJ j- J3 3 •a cd 73 c PCL o •3 .5 o 3 JS n OH J on CO 03 03 3 U co u J3 a u 03 i Ml NO CO m O CN m o ON CN m r- in m CO CO ON ON NO CN ON o oo oo oo m © m ON r- o ON o d © O o © © o © NO m NO in CO oo CO *d- r- o in CO ON CN 00 m NO ON ON ON ON ON m OCON r- OCN co CO oo o CN oo CN 00 o o 00 o © © 00 m © in ON ON P- o © 3 d © o O o © © © © o © © NO OO CO CO NO 00 m NO r- in ON ^H NO CN oo ON ON ON r- CO o ON O>Nn m OCNO oo 00 00 00 CO 00 d O d d d ON oo NO d ON ON d d d d d d d d d d i m 00 o ON CN ON NO ON d ON in o ON ON NO 3 CN en o d d d CO •* ON 00 d © d d d d OH CN CN CN CN CN CN NO CN - CN o BH CN NO NO m m NO o o CN •a _ — T3 — -a 73 73 73 73 1 cd cd S3 1 b b b b cd cd 5 J3 3 ii B B B > JS 3 3 B I-H Ii cd B O O Bc d Bc d Bca Ccd 3 3 J= CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ cd cd cd cd cd cd cd cd cd cd cd cd cd cd C3 cd cd cd cd cd CJ CJ CJ CJ CJ cj CJ CJ CJ CJ CJ CJ cj CJ CJ CJ CJ CJ CJ CJ o o CJ y CJ o CJ cj o CJ CJ CJ cj o CJ CJ CJ CJ CJ CJ Bcd ncd cBd cBd Bcd Ccd acd Bcd cBd Bcd Bcd Bcd Ccd Bcd Bcd Bcd Bcd Ccd cad Scd E E £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ T3 =CD3 a -i—i a o s o C3 3 « 3 3 3 •a C c a to _g o Tt ON —-, NO Tt o o Tt ON no m NO OO mNO OuoN ON ON ON NO Tt o ro CN co OO © d od o d d od o od o 0d 0 d d d ON d ON r- Tt NO NO Tt t> NO o OO 0N0O 0o 0 Tt ON m m 0N0O Co N NO nOoN m ON ON ON ON ON NO Tt d d rOoN CN od o oo d CN d s d d 0d 0 d d 0d 0 0d 0 do o d 0d 0 d 0d 0 0.9 8 0.8 6 OCN OCON — NO m m OO NO If! d d 0.9 4 Nd O NO oo Od N 0.80 8 0.98 6 0.98 3 0.97 1 894- 1 -0.8 4 -0.9 7 O ON i CN -H f~- 1 ON o t— ON 00 o 0.6 8 d Od N 0d 0 .926 - 00 0.76 5 o d 0.751 - Tt i»o CO CO "JO CN —-1 Oa in NO o NO Tt NO Tt NO NO CN CO -H _I CN T3 73 73 73 73 S3 7b3 73 73b 73b 7b3 73b 73 73 b b b b •ss S P CbO £3 c3 c3 ca CO nat i Z| nat i E nat i nat i nat u ECO z Z z Z CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ ca ca cCaJ cCaJ cCaJ ca cCaJ cCaJ cCaJ cCaJ cCaJ cCaJ CJ CJ CJ CJ cu cu CJ CJ cCuJ CJ CJ CJ CJ CJ CJ o o o o o o CB o CC) Bca Bca cBa cBa Bca Bca Bca Eca cBa Eca E cEa Eca cEa Eca Eca Eca Eca Eca Eca t-H £ E E E E £ £ £ £ £ £ £ £ £ £ £ E £ £ E TC3U _C3 a o o CS C3 ca s -C3 •a ft. a a o s <3 -8 T3 R C3 3 3 CO s a Cj CJ Cj cu cj «D C-5 .R a. OH "8 3 3 3 3 3 3 3 3 CM R R R R R R R S. R < cu a. a. a. a. 3 ft R CN aos OS Os Os m Os CN OS CN OS o CN 00 00 OS o o 00 oo OS Os CO ^ as OS o o OS OS o o Os 00 oo 13 CN Os Os 2 o CN •3 U I 1 O J3 x =8 «8 < < •a 2 2 =8 00 •2 o 5 a 00 00 cd C8 o o c a J3 J3 _*e O 3 2 2 CO o o CO CO OH Q Q O CN m CN SO CN o Os OS m in CN OS SO 00 d d oo Os Os 00 00 d d d d d 0.9 6 m CN Os 00 in Os 00 SO 00 CN o o CN 00 Os OS in u-> O^*s Os 00 Os 3 m OS SO CN o CO 00 d d oo OS d d OS d OS oo Os OS OO Os Os 3 d d d d d d d d d d d d so oo SO oo CN CN CN CO SO CN Os Os Os CO SO Os OS 00 CO r~ SO SO O SO Os OS in d o d oo OS Os d OS OS Os oo Os 00 d d r- d d d d d SO d d d 1 d i d oo 1 * m A *d- oo CO r- Os . S0O0 d mr-- min OS OS CO OS Os oo d OS d d d CO O CO PH , CN —i —' CN -H PH -d- m _ -o •a •a __ __ T3 1 S3 S3 73 13 "c3 i3 Cd S3 ca ca b b b b b o J3 J3 p p p p J3 3 3 I cs Bcd Bcd Bcd B CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ cCaJ CcaJ CcaJ CcaJ CcaJ CcaJ CcaJ CcaJ CcaJ CcaJ CcaJ CJ CJ CJ CJ CJ o CJ o CJ CJ o ca ca ca ca ca ca ca ca ca ca ca tS 1 B B B c p b a S3 a a c a a c c c c c OH E E £ £ E E E T3 CO a -4—1 oC cj R R cu cu 5 S "3 i; i ft CJ C3 •8 do C3 3 3 a Oo 00 R R ^3 -ft -ft co 00 3 3 a O a 3 to to ft, ft. I to to to a to •§ •§ CD 3 3 3 3 3 3 3 •§ 3 a R a a R R a R a R R R 3 3 a. a at co R R £1 R R 0, < a. OH O•cSt OS in o -a •ct O OS 00 CJ OS Os o OS OS o —1 Os CN OS so OS OS 2 a oo OS o a cu _L. OS •S <*! co £ •a =8 ca c o bo U CU ^ -a S 2 •c OS m o d Os d m m so m CN CN t- Os •ct OS o OS m in OoS OS © d •* © d d d d d o •cl- CN o o O so m 00 m m 00 o o CO CN o in CN CN SO so r- CO oo CN o •o Os OS OS 00 OS OS OS so CN CO in OS 0©0 d d d d d d d d d d d d d d d o so CN OS so CN m m m CN r- CN CN So SD OS CO CO m oOs rO-s OS p •*Cf OS Os CO CN oOS d 1 © d d d d• . d m^ Os d d1 d d d CN oo COi CO m r~ o OS OS om CN r- d d d d d PH -ct Os CO CN —1 O PH SO so o oo CN Os ' r- CO OS CN V V •a -o •a H S3 S3 S3 "ca c3 "ca ~a "ca "ca "ca •3 |3 |3 > J3 J3 Jo3 JO3 j= ca Si a u< ca a ca ca ca ca ca z c O o O O c o c s s B o c CJ CJ CJ CJ CJ CJ CJ CJ ca ca ca ca CO ca ca ca CJ CJ CJ CJ CJ CJ CJ CJ -a cCaJ cCaJ cCaJ cCaJ coa cCaJ coa CcaJ o <3 (3 C3 (3 (3 n a c a. 3 u u u u u T3 CD =3 B ca -*-» ca B Co to CO to to to to O cu cu cu cu cu cu cu CA o KJ KJ KJ KJ c« s o Scu S scu scu cu Kcu cu cu •»! cu 2 u ca s s 5 s s s s s s a. « H a <3 a a a a ua S cu 00 00 00 00 00 00 cm cj cj cu a SA sco t3o 3 3to .t3o .t3o 3to ca si •« ca T3 IH B o O o o O 0 o IH o Cj Cj a. a. ca 00 CU CD •« •« <3 cu IH •5 CH •a •C3 -a •a 00 > 0 « C3 CU 5 DH acu 3 3 c3u c3u c3u c3u 3 3" .« 3- C3 > s Si to ctou ctou to to to to < 5- < < a. a. a. a. a. a, a. g 3" g g Tt ON ON ON ON ON ON ON ON ON r- CS J2 ca M M ON Tt Tt •* © ON ON ON ON ON Tt ON ON ON o o o o cj ON -OMN -M -M ON NO o •^J O ON o C3 C3 13 ON cos p» Tf 00 m in ON oo . NO ON NO m Tt en in 00 r- ON ON NO CoDo doo dm doo eCnO oNOo o 0P0- pcs d cONs d OoNo ON odo pON- d d d d d d d d P- Tt 00 in CO oo ON oo NO ON NO m Tt en p- NO 00 p~ 00 00 00 ON ON ON oo m iTnT NO s ON 0P0- cs cs ON oONo oo pON~ S d d d d cs p d ON d d d d d d d d d d d d ON ON 00 NO ON ON Tt CO 00 P~ en NO eo CO p- 0c 4 NO ON 00 ON ON ON ON oo O P- OO oo r-i °N o cs m OT-M d d d Tt CO Tt d d• d d d cs d ON 4 ° s ON ON ON ON NO d ON Tpt- ON d d en d d PH es -M i—M cs cs -M Tt CS Tt -M o es PH Tt Tt m NO in CO P- in o Tt P~ ON 00 CO 73 73 7a 73 7a 73 73 73 73 73 73 73 73 73 73 ca 7l3H 73 H u j3 3 3 3 ]3 |3 jj •3 c3a 3ca ca M M ca ca ca ca ca C8 ca c B B B E B E E E E E Ec3 E 1c3 E E E E cj ej CJ ca ca ca CJ CD CJ CJ CJ coa ca ca CJ CU CU CU & & e- 8 cCaJ cCaJ cCaJ cCaJ T3 CJ CJ CJ CJ ca ca ca ca cCaO cCaO cCaO cCaO 2 e o o o O B B -c JJ H. a, E E E E E E ea, O o o o o o Mi l Mi l fa CQ 03 Mi l Mi l (3 3 03 ca c-1 T3 CJ .g •I—* a C3 C3 o ft. e a. cj Cj -if .-i0f -C3 cu a. a. s ca a I, T3 o s •3 c? a Cj ft. C3 es a 5 3 3 CJ Si Cj CS -a C3 00 00 Cj 3 Cu CO CM -CCO 03 3 a 03 03 c3 < in in Ov Ov cs O o CO o Ov VO vo vo Ov VO CO •B Ov Ov Ov Ov oc s Ov Ov Ov Ov Ov Ov Ov ca OV •"H =8 3 «^ ~ ca CO <3 « CL) *- ^ cu cu cu >, CJ o> . CJ CO CO 3 CO CO CJ £i a i33 CQ X 3 s? 0, X 55 CO CO m CN m CO VO VO Ov OV VO p *ct o Ov p d 3 r- Ov d d Ov d d CO CO 00 00 in CO vo VO VO OHO Ov r- in CN Ov Ov p -ct o 00 OV OV Ov Ov 3 p d d d d Ov so d d d d d CO H CO 00 I—H oo m cs m CO VO vo VO OO Ov r~ m CN Ov Ov VO o "3" o Ov Ov OV Ov o p d d d d 3 r- d d d d Ov d 0.9 3 0.9 5 1.07 7 1.00 6 1.03 1 0.98 6 0.95 1 0.92 1 0.99 8 0.64 2 0.96 7 1-0.9 2 •1.02 5 •1.30 4 0.59-0.6 7 d 1.03 - 1.008 - 0.986 - 0.989 - 0.97 - CM cs cs cs - cs - - O - •cf •* m o •* CO m CO o m Ov cs cs O 13 a 73 13 "co "co "co 13 "co 13 13 a 13 CA b b b b b b b b b b > b nat i nat i nat x nat i nat i nat i nat i nat i nat i nat i nat i nat i nat i CJ CJ CJ CO CO CO CJ CJ CJ CJ CJ CJ CJ CO CO CO CO T3 CJ CJ CJ CJ CJ CJ CJ •e CO CO CO o o o CJ S3 o J3 CO CO CO CO o O. e t ts J3 XI J3 JD J3 CO a CO CO CO CO 3 CO 2 2 2 LL, LL, LL, LL, •a S 2 C< L3O •a CU 3 2 _g S o K CJ a s cu 00 •§ cu -B Cj 5 £ 5 K o •5 •3 •3 5 H 4 5 -S -o5 a •3 .« .35 -a <3 <3 0 •S O ~ cj •SP S a u .3 -« e cj « -S SJ -5 a <3 CJ o a J3 ,t-> 3 3 OH 3 < 156 Appendix IV Western red cedar foliage DNA CTAB extraction protocol (for 12 samples). 1. Chill mortar and pestle in -20°C freezer overnight 2. Preheat 2 x CTAB buffer at 60°C (or more 65-70°C) 2xCTAB buffer 0.1 M Tris HCLpH 8.0 for 200 ml: 20 ml lMTris 1.4 M NACL 56 ml 5MNACL 0.020 M EDTA 8 ml 0.5 M EDTA 2% CTAB 40 ml 10% CTAB dH20 76 ml dH20 2% P-mercaptoethanol 4 ml p-mercap 3. Grind 0.25g of foliage using liquid nitrogen and place in 50 ml tubes 4. Add 15 ml of CTAB buffer per sample, incubate 30-45 minutes, swirling every 10 mins 5. Spin on table top centrifuge at 4000 rpm at 0°C for 10 mins to remove most leaf debris 6. Transfer supernatant to new tube. Add an equal volume of chloroform:isoamyl alcohol (24: T) to the extract. Mix 15 mins in rotating machine. 7. Transfer top part to new tub and spin 10 000 rpm at 4°C for 10 mins (or more 30 mins) 8. Remove top phase and transfer to new tube. Add 2/3 volume cold isopropanol. Mix gently and spin down DNA Pellet. Pour out isopropanol and air dry. 9. Wash in 20 -25 ml wash buffer of 75% ethanol for 120 ml: 90 ml 100% ethanol 0.01 M amonium acetate 0.24 ml 5M amonium acetate H20 30 ml H20 10. Spin, pour out, dry 11. Redisolve DNA in 500 |xl TE at 4°C 12. Add RNAse at a concentration of 10nl/ml (RNAse 10 mg/ml stock) Incubate at 37°C for 30 min 13. Add Proteinase K to a final concentration of 10^1/ml Incubate at 37°C for 30 min 14. Add equal volume of phenol and chloroform (1:1), rotate for 15 min. Spin down at 8000 rpm at 0°C for 10 min Transfer supernatant to new tube 15. Add slightly less than equal volume of chloroform:isoamylalcohol (24:1). Mix for 10 Spin down at 10 000 rpm at 0°C for 10 min Transfer supernatant to new tube 16. Add NACL(5M) of a final concentration of 0.2 M. Mix well. Quick spin Precipitate with 2 volumes cold ethanol. Mix gently Store in freezer for 30 mins or overnight. 17. Spin down at 10 OOOrpm for 10 min 18. Remove supernatant. Wash with cold 70% ethanol and air dry. 19. Redisolve in 500 ^1 H20 158 Appendix V Twelve western redcedar DNA sequences for which primer pairs where designed and that amplified scorable and variable microsatellite loci. The position of each primer is underlined. All the sequences have been deposited in Genbank. N, H, W, M or V, base not known. Locus Tpl Genbank accession no. AF245205 CCCAAGAGTTAGTATATCTCTTTGTCATGTGYBNSATTTCCCTTGGTTTTTCCTTGCTT GGGGCTTCTCATAGTGGAAATATCTCTTGTGGGTGCATGTATGCTCCATTAAATCCA TTTATCCCATTAAGGCATTTGCAATCAATTTATTACGCGGGAATACACTCTTCTCCTT TTTTCACATTTTGCGCGCACGTACACACACACACACACACACACACACACACACACA CACACACAATGATTTAATACATTTTCTTCTGTGGATGGATAAGGAGACAATGCTTTC GTTGGACTTATCTTTACATTGGCATTCCTTGTGCATCCTTAATCAAATTCCTTAACAA AGACAATGCTTTCGTTAAAGACCGATCCTCCTTTTCTAGATACCTTTGGAAACTATGT TTTCCTTAGGACCAGTCGCTTTTTTTGGATCCCTTAAATAAGACAATGCTTCCCTTAA GGACTA Locus Tp2 Genbank accession no. AF245206 GGTCTTCTAAGTGCCTTAWAGTGTACTWGGACATGTACTAWGAYATGTAATATTTT AACATCTAGCTTTGTACATGTHAATAACATAAAAAAATACATAACTYTHAAACATAT ACATGTWCATGATTGTGTGTGTGTGTGTGTGTGTGTGTGTGTKTGTGTGTGTGTGTG AGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAG AGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAGAG AGAGAGANATTCCTAATAAATTTACTAACTTTAAAAGAAAATATTTTTAATTTCTAA CATTTCTACTANGGTTTTCTAACATTTTCTAACATACTCAGTTACACTNTATTATAAC ATTTTGCTACATTCTAATTACAGTTTTCTAATGTTTTCTAATGTGTACTAGGTTATAAT TTTCTA Locus Tp3 Genbank accession no. AF245207 CCAAATGCGTATAGTTAGTTWAAAGACCAAATWAANRATAGGGATCAGTCTCAAG ATCCACTAAAAAATATTATAAATCTGTGTGTGTGTGTGTGTGDGTGTGTGTGTGTGT GTGTTGTTTTCGTAANTCCAAAANNATNTTNTTNTTCAAATCAANAACCCCACCCTTT TNNCCTNCANCCCTTACTAANATAATTGCATTTCCAAGTTATTTTTATAGGAAATAA ACCTATGCAAMAGTACATTTATTAACTTAACAAAATTATGATACATTAATTAAACTA ATTTACCTAAAAGACAAAATTATCTGCGTCTTACAAAATCGTATGATCAAAGTCTTT TTTAACTCACTGTAACATAGGTATTGAGGGATGTTGGAATGAGTTTCCAGGACACAG GAAAAAAAA 159 Locus Tp4 Genbank accession no. AF245208 CCCATCTTGCCACTTATTGTAACTTGTATAACCACTACATCATTGGAAAAACAACTG ATAATTTTTTCCCATGCTAGCATTCCTATCAAATTAACTAACATCTTGGCTCAATACA TATTTTAAATATCGCATTCTTATGTGAGTTGATTATCGCTCACTGATTATATTGGAGT GGAGTTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTATTAGTGTTGTTAATATT ATATTGGAGTGACCCTGACAAATCTCCAGGATCTGATGGCTTCCAAGCCTTCTTTTTA CAAAAATGTTGGGATNNNCATTGGGGCTGAATTATGTGCAAGCAATGCG Locus Tp5 Genbank accession no. AF245209 CCAGAAAATATTGATGTTTATAGCTTCGATCAGATRAAATGGTCATAATAATATGAG AAAATAGATAAATAATATGTGTGTGTGTGTGTGTGTGTGTGTGTGTGAATAATGTCA TTACTATTATATATATTNATATACATGTTTAACATTATCCCTTTATAATATTAAACTV YATAAACATCAMTGAATTGATAATATYAAATAATATTATTCTACATTAATATATATA GTAATAATGATATTATTATATATATATCAATACACATCTACAATGTATAATGTATATT ATATTTGATTAAATAATATTTAATAATACTATGCGATACCGATTAGTATCGGTCAAT ACATTGATATTGATTTAGTTAAACAG Locus Tp6 Genbank accession no. AF245210 ACTTAATAWATGTTGTAACATTATGACAACCTACTAAGAACATCCYAATCATAAAA ATTGAACCAACACACAAAACCGCCCATGTAAAAGGAGGAGGAAAACYAGGAGATC ACAAGCAATGAATGAATAAGCAAGAACATATACGCGCGCGCGCGTGTGTGTGTGTG TGTGTGTGTGTGTGATATGTATATGTATATGTATATGTATATGTATATGTATATGTAT ATGTGTGTGTATATATACATATATATACACATATATACACACATACACACACACGTG TGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTACACATACGTGTGTATAGCCCTA GTCCG Locus Tp7 Genbank accession no. AF245211 GTCATATACAGRGTTGCCAATTTGGGTAAMATCAAACCTTATCAACCTCCCATGGTG TGGGGGTGGTCCAACACATGACTAGGACCTAACTMAGGACTTTTGAMATATAACTT CTAACTAAACACACAAAACAACAACTAAAAAGGTATAACCATGGCGTCAAGAACCT AGGKTTATAACTTTACCTTAGGTCCTTCGGCTTACATCAAAGACCCCAACACACAAA CACACACAAGGGTGATCTAATGGATGCACCTTATTAAAAACTCAAMAGRCAGCAAA ATTAGGRGTTTAGGRGTTTTAGGTGGKGTCCTGGCACACACACACACACACACACAC ACACACATATAGATCCCGCCTNGGACCCCTTNGGACACATTAGGGTNTCTTAAGACC CCTTAGAACATACTAATCANCGTNGGATACACTAATGGGTNTTAATGGGTTCTAAGT GGTCCTAGGGACCCTCAAGACCCATTACGACCCTTAAGACCTATTAGGACACATTAG GAC 160 Locus Tp8 Genbank accession no. AF245212 ATATTAGGTTAATAGCTAGTTGAWATTTTTACGTHAAGTTTACTAATCTTGCACACA CACACACACACACGCACACACACACACACACACACACACACACACACACACACATA TATATGTTCTNACAAATTTTTGATATTAGGTTGATAGCTAGTTGAAGTTTTCACGTCA AGTTGACTAACCCTTATTTNNTATTTTTTATTGCCACTTANNNGCATAGTTTGG Locus Tp9 Genbank accession no. AF245213 CCAGGDCAGGTTACATAGTTTCTAGAAACCCAATACAGACACCAGTAAACAATAAT TTCCAGATCTTCTATATTTYAAGATAAGTGTTTCATTTCATCCTAATCTTTCAGATCT ATTGTTTTACTTTATAATTAAGAAATTAGACCCCCTTCCCCTCTTTATCCATAACCAA TTGATTCTATTTTTMACTGATCATCTCCTTTTGGTTAATCAATTVAVNTCTCCTTGTCT TGGATTTGGAGATAATATATATATACAACATTTTCCATACATAGAAAACAAACACAC ACACACACACACACACACACACACACACACACACATCTTATAGAGTTTAATAACAT AAAACTAAGTATATAAATAATGTAATCCAAAGGAGTAATNCGTAAGATTGTAAAGT ATGATATGTACTCCATTGAAAAAAAAATTCTATGTTTTCTAGTGATAATGAGACTAC TJTCCGTAANAANNCCNNNNAAAAAGTTTGAACAACATGACCACTACAAGG Locus TplO Genbank accession no. AF245214 GGTCTTGTTTATAGTTGTGTCCATTCAGGCATAAACATATATGTGTGTGTGCGTGTGT GTGTGTGTGTGTGTGTGTAATGAAAAGATATAGTTGAGTGTCGAAATAAATAAAGA GAAGAGATATAGAAGAGCAGTTTAAAGATGTTAGTATGAGATAGAGCCCTAAAAGA AGAJAAGAGCATAAGTGCATAATAAAAACTGATCAGACATTTGTGTCATATATAGT AGTTGACTTTTGGATATAGTGCATATCTATTCTCAGGAGGTTATAGTCTTCTTTCTTA TTGAGCAGTGAGCTCTTGGACAGTGAGTCCTCACCCCTAACAASGCTGTTTGTAAAA GACTCCTAATAGGGTCAGG Locus Tpll Genbank accession no. AF245215 GAATTCGGACTACCTGATCCGCTTTGATGGGTTGAATAGCAACTATGTAGCTTGCTA GAACTTCCAATGCCATATTCTTTTGCTACACTTTTNCCTTGCATGANTNATAGTAAGG NATCTCCTTACTAGGCTCTCTCTCTCTCTCTCTCTCTCTCACACACACACACACACAC ACACACACACACATTTATCTATCTNTCTCGAGTGATGCCTCTTATCTTGGTTTTGGAT TTGATGAAGACTAGTCCGAATTC 161 Locus Tpl2 Genbank accession no. AF245216 TATGTAGCTAGTTTGACCGAAGTTTGACCCGCCATATCTCAGACGTATGGACTCCTC TTTCACCCGTCCAAGCTGCATTGAATCCATTTTCTCAAGAATTTTAACTTGGTGTTAA CTT ATGG A GT ATTGG AG AT ACTTC A ATTTT A AGT A A A A AT ATCCCCCGGATCATTAA GGGCTCTATCTCATTCTGATATCTTATACTGCTCTACNNNNNNNNNNNNNNNNNNNN NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNTGTGTGTGTGTGTGTGT TGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTTTATGCCTGAATAGATGCGACNAA AAACAAGACC CO u CN ej CQ CQ C14 C15 — ro< wo \ in CN—' d